Look at GPU benchmarks. Now look at CPU benchmarks. The lower numbers are your fps estimates. Not that hard.

Technically you want infinite fps.
You also want the fps to be a a multiple of the refresh rate to avoid screen tearing. With 60 Hz double or even more was easily doable. With 120/144 double was the only realistic multiple. I don't see it happening with 240.

G-Sync/FreeSync also change this a bit because now you can avoid screen tearing without having to hit your fps cap 100% of the time. 240 Hz Sync with ~150fps obviously isn't really better than 144 Hz capped at 144 fps. So you still want to get as close to the refresh rate as possible, but going above it isn't that useful anymore.

tl;dr
No G-/FreeSync: x Hz with less than x fps will look like shit because of screen tearing.
With G-/FreeSync: >x fps do nothing but <x fps still look smooth.

Don't worry, for the next 2 years they'll play both their home and away matches in a city less than 5500 miles away from the city they're representing so the home crowd should easily be able to make it there to cheer them on.

As mentioned by #6 also Taimou for Dallas.

Now you might think that means 4 for Houston and 3 for Dallas, but you'd be wrong.
http://etf2l.org/forum/user/1630/
That's cocco.

So Covoo you need to decide: Dallas (4/9) or Houston (4/10)?

No, wifi is actually illegal above an elevation of 1000m if there fewer cows than humans within the area.
The reason is the wifi leaks into space and attracts aliens. If they can't be tricked into thinking that cows are the dominant lifeform things might get nasty.
No, you can't just use something else in Mongolia, we've committed to the cow ruse everywhere else so the aliens would notice.
That means no wifi for Mongolians.

Basic model of your internet being routed:
Green: to Europe
Red: to China
Purple: China alternative route
Blue: from China to South Korea/Japan

#31
Meltdown means any website with javascript could read anything, including passwords, from your RAM. Any program could obviously do the same.
Basically
https://twitter.com/misc0110/status/948706387491786752
It shouldn't affect gaming much but either way are you in a position where you can tolerate the risk? Because that thing is going to be spammed since enough people will be dumb enough to not apply the patch.

Is the room colder than 12°C? Because no amount of fans can cool anything below room temperature.

You should probably use a different program.

https://i.imgur.com/klqPKP5.png

Seriously though, those are idle temperatures, not load. How did you check them?
PSU shouldn't be a problem. Running out of power crashes the pc, it can't magically choose to slow down.

PCPartPicker part list / Price breakdown by merchant

CPU: Intel - Core i7-8700K 3.7GHz 6-Core Processor ($388.89 @ B&H)
CPU Cooler: Noctua - NH-D15 82.5 CFM CPU Cooler ($87.95 @ Newegg)
Motherboard: ASRock - Z370 Extreme4 ATX LGA1151 Motherboard ($189.88 @ OutletPC)
Memory: GeIL - EVO SPEAR 16GB (2 x 8GB) DDR4-3000 Memory ($174.99 @ Newegg)
Storage: Samsung - 850 EVO-Series 1TB 2.5" Solid State Drive ($289.00 @ B&H)
Video Card: MSI - GeForce GTX 1070 Ti 8GB DUKE Video Card ($514.98 @ Newegg)
Case: Fractal Design - Define C ATX Mid Tower Case ($79.99 @ SuperBiiz)
Power Supply: SeaSonic - FOCUS Plus Gold 550W 80+ Gold Certified Fully-Modular ATX Power Supply ($72.89 @ Newegg)
Total: $1798.57
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2018-01-05 07:19 EST-0500

Obviously you can get an i5-8600K instead, 32GB RAM, a 1080 or even a 1080 Ti if you downgrade some other things.
The case is slightly smaller and just an example of what you could get. It doesn't have external 5.25" slots (or any 5.25" slots really) so if you need an ODD you need to pick a different case.

Since you want it to be somewhat portable I'd actually downsize to at least micro ATX. Mini ITX is usually more trouble than it's worth.
PCPartPicker part list / Price breakdown by merchant

CPU: Intel - Core i7-8700K 3.7GHz 6-Core Processor ($388.89 @ B&H)
CPU Cooler: Noctua - NH-D15 82.5 CFM CPU Cooler ($87.95 @ Newegg)
Motherboard: MSI - Z370M GAMING PRO AC Micro ATX LGA1151 Motherboard ($168.56 @ Amazon)
Memory: GeIL - EVO SPEAR 16GB (2 x 8GB) DDR4-3000 Memory ($174.99 @ Newegg)
Storage: Samsung - 850 EVO-Series 1TB 2.5" Solid State Drive ($289.00 @ B&H)
Video Card: MSI - GeForce GTX 1070 Ti 8GB DUKE Video Card ($514.98 @ Newegg)
Case: Fractal Design - Define Mini C MicroATX Mid Tower Case ($69.99 @ SuperBiiz)
Power Supply: SeaSonic - FOCUS Plus Gold 550W 80+ Gold Certified Fully-Modular ATX Power Supply ($72.89 @ Newegg)
Total: $1767.25
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2018-01-05 07:30 EST-0500

That case isn't that much smaller so there are much smaller options.
The mobo will be slightly worse for overclocking but not much. You can check the other differences in features/connectivity yourself, you probably won't need most of it anyway. Carbon would be cheaper but 2 RAM slots aren't a good idea since you might want to add more later.

Why a new 200R? Either way at the price Corsair wants (>50$) no dust filters is just inexcusable.
Ok.
Ok. Mainstream SSD then, nothing too fancy.

Yes.

Frankly at this point I can't say how much either will be slowed down (Spectre affects both but the fix should be cheaper).
Zen isn't really going to get past 4 GHz so it'll definitely still be slower single threaded in applications that aren't affected much, especially since Coffee Lake could be overclocked to ~5 GHz, but on the other hand Coffee Lake is really going to suck now for transfering lots of small files.

I forgot to mention: VGA will be difficult, any recent GPU should support only digital.
Also: Do you want a Quadro/FirePro? https://www.pugetsystems.com/labs/articles/Why-you-should-use-a-Quadro-video-card-in-Solidworks-2016-751/ Obvious shilling in the title because they want to sell Quadros but it's worth considering for the drivers since you don't need more than a midrange GPU for everything else at 60 fps or more.

I don't think anything except TF2 will be CPU limited anyway so I'd pick the CPU based on Solidworks performance vs willingness to put of with some things being much slower on Coffee Lake.
https://www.pugetsystems.com/labs/articles/SOLIDWORKS-2017-Coffee-Lake-CPU-Comparison-i7-8700K-i5-8600K-i3-8350K-1053/
1800X and 8700K would end up costing roughly the same, 1700(X) overclocked would be about the same speed as the 1800X but cheaper.

I wasn't sure where you were going with pointing out that JITs exist outside of the kernel.

Well I wanted to point out the difference between the "shit's on fire" situation that is leaking kernel data from userspace with default configs (Intel) or non-default config or a more sophisticated variation (AMD) and the eternal dumpster fire that is browser security with javascript. Because the former will be fixed soon, the latter will only be fixed when javascript etc. are running in a seperate process. That's the reason why that feature exists.
On second though if this is what it takes to make site isolation the default then sure, go ahead and warn as many as you can.

They don't have to be but that doesn't matter.
For the attack to work the executed code needs to be able access the data you want to leak. So if you're not in kernel mode an attack on the kernel will fail because the data is never accessed either a) because the permission check fails (AMD/ARM) or b) because the fix for Meltdown has already flushed the TLB and the adress isn't known.

Sure you can use it to gain access to data in your own process but that's a bit redundant. If your own program contains malicious code and uses it to leak its own data to itself, well, you've achieved nothing.
If you're running untrusted javascript code via JIT in the same process as some sensitive information you might want update your JIT to use conditional bound checks. On the other hand you're running untrusted javascript code via JIT in the same process as some sensitive information so you deserve everything that happens to you.

Spectre isn't really fixable on an architectural level. Even in order cores are affected if the pipeline is long enough.
Ice Lake should have taped out by now. It's a matter of steppings and yields now. Changing when permissions are checked (before the loaded value appears vs at retirement) affects basically the whole CPU and at the point that Intel learned about the flaw they were most likely unable to change it. There's also the matter of AMDs patents so it might be quite a while before they fix it. If I'm correct it would be quite difficult to even get it into Tiger Lake (Ice Lake refresh). If they can still get it fixed on Sapphire Rapids, the architecture after that (2020+), which they are already working on because development takes ~4 years, then they might consider themselves lucky.

That's just the usual trifecta of
"It's not a bug"
"Everyone got it"
"It's not as bad as you think"

For example

actually means "yes you can read kernel data".

Right now the status is:
Variant 1 (Spectre/bounds check bypass):
Patchable, use conditionals instead of branches.
Relies on kernel-mode (so in the OS) interpreters or JITs executing the code so as soon as those are patched it's fixed.
May not even work with default configs on AMD.
Should not affect performance noticeably, especially if you're not using anything that's affected anyway.

Variant 2 (Spectre/branch target injection):
Patchable, don't use indirect jumps or prevent speculation on them or if possible invalidate the branch predictor on a context switch.
Depends on the specific branch predictor (CPU hardware). Intel Haswell has been proven to be affected and ARM says it also affects all Cortex cores. AMD claims theirs won't be vulnerable to it, but they are the only ones that actually know the inner workings of their hardware so it's difficult to verify.
The brute force patch Linux will use for now will affect performance, but I don't know by how much. Future patches might reduce the impact.

Variant 3 (Meltdown/rogue data cache load):
Patchable, flush TLB on every context switch.
Depends on the specific architecture. Might affect anything newer than the Pentium I from Intel (except Itanium and first gen Atom). According to ARM themselves only the Cortex-A75 is affected. AMD claims they are immune and so far it looks like it.
This is the one everyone is talking about. The performance impact depends on what you're doing. Pure number crunching won't see a noticeable change but for example file accesses will slow down rather significantly.
Unlike Spectre where multiple patches may still happen to close new security holes or to lessen the performance impact this will be more or less final. Unaffected CPUs may be added to whitelists later but if your is affected then it's going to stay like this.

It's not that easy, you still need the malicious code to be executed in the first place.

And I doubt the NSA was involved at all. They could never pay them enough to compensate for the effect this will have on their sales.

It was simply a design decision of "page faults will be thrown at retirement anyway so why bother checking?".
Intel probably never changed that because AMD patented the checking iirc and AMD only does it because it prevents bullshit data from being cached and is rather cheap, not because of security concern.

Correct.

In theory it shouldn't change much for TF2, although TF2 does weird shit with the TLB too so it could be either better or worse.
Generally games won't be affected much.
https://www.phoronix.com/scan.php?page=news_item&px=x86-PTI-Initial-Gaming-Tests

In general though it is rather annoying and fucks filesystems completely.
https://www.phoronix.com/scan.php?page=article&item=linux-415-x86pti&num=2

Don't confuse Spectre and Meltdown.

Meltdown = fix is coming, but causes slowdown = Intel only
Spectre = everyone is fucked, but difficult to exploit = shown to work on Intel, only works on AMD within the same process or with specific non-default settings = no fix yet