"CPU Throttling" will turn red and change to "CPU Throttling - Overheating Detected!" and a red line should appear in addition to the yellow cpu usage line.

http://i.imgur.com/4EzHSuA.png

CPU-Z has never shown temps, not sure what Shoosh thought.
Not sure why higher fan speed should affect power consumption significantly either.

Use AIDA64, it'll show temps and throttling.

So pretty much standard?
I haven't gone over all prices and specific parts yet since I don't know which shops he'll be ordering from.

i5-6500/6600 if he thinks he'll need it. I don't think an i7 would do anything for GW2.
Some mobo
8GB DDR3 or DDR4, DDR4 still a bit more expensive in NZ
850 Evo 120GB
WD Blue or Seagate Barracuda 1TB
R9 390 or GTX 970, depending on the shop->price
N200, maybe PS08B
CX430, 500B, CX430M, HCG-520 or HCG-520M

Nothing terribly exciting.

I'm only going to look up specific parts and prices once it's clear when he's going to order. No point in including any deals that'll be gone by the time he orders.

Budget shouldn't be a problem so let's focus on performance.
i5 should be enough.
Define decent GPU.
SSD size? HDD?
µATX ok or SLI/Crossfire or lots of PCI(e) cards planned?

I'm away for 2 days and this happens.

Step 1: Ask for advice.
Step 2: Ignore advice.
Step 3: ????
Step 4: PROFIT!

This is the first thread in which I mentioned the ST31000524AS and talked about it being ancient and not worth buying.
What does he buy?

Case is garbage tier
Cooler is garbage tier
mobo is garbage tier
PSU is might-catch-fire is garbage tier

I'm taking any bet that he went over budget aswell.

Needs more POWERRR!
https://www.youtube.com/watch?v=bbLshnfu0wY
https://www.youtube.com/watch?v=Ee5evlN8Bbs

#6
So it's an act? I probably still wouldn't fry my fingertips for the internets.

What about BF4? It's Crysis 3 and not Crysis 2 btw, but don't let facts get in your way.

http://www.pcgameshardware.de/AMD-Radeon-Grafikkarte-255597/Specials/Radeon-R9-390X-Test-1162303/
10% faster.
If it were the exact same card, which it isn't, and it was just a 10% overclock you'd still be comparing an oc'd card and a card on stock clocks. Also the 780 Ti still doesn't appear in those benchmarks, so you can't compare them anyway.

Are you still trying to invalidate a Crysis 3 benchmark with Crysis 2 and CAPS?

Same can be said for an i3.
http://www.techspot.com/review/991-gta-5-pc-benchmarks/page6.html
Nevermind that the FX 8 cores compete with i5s. They never made sense for games and it won't get any better the older they get.

Sooooooo, you linked Project Cars benchmarks to show the 290X is faster and when the benchmarks don't agree with you anymore AMD cards are suddenly "known to have issues" in Project Cars?

My point exactly. If you compare reference cards the 290X is crippled. But if you want to compare overclocked aftermarket cards then do so, don't compare an overclocked aftermarket 290X and a reference stock clocks 780 Ti.
If this were about price to performanc it'd be difference, but you said the 290X is faster than the 970 and 780 Ti and the 780 Ti is slower than a 970 and both of those statements are false.
Best case for a 290X is about 1200MHz. A 20% OC.
A good 780 Ti will do 1300MHz. A good one, not a great one. That's a 40% OC.
Now tell me which magic will make the 290X 40% faster with a 20% OC so it stays ahead of the 780 Ti. If it even was faster on stock clocks.
If you don't want to compare custom overclocks we can compare factory OCs, but it won't look any better.

Thanks for the validation, I really needed that. Oh wait I don't. I always do my research before I buy anything. That way I always get good cards. I own both AMD and nVidia GPUs. I live on both sides of the fence, my grass is always green, metaphorically speaking.
Please tell me what CCC is. I really don't know. I've never seen a computer in my life. I don't even know what they're supposed to do, I just built and sold them.

I blame the 8350 on the 680 build, since the 680 definitely isn't Fermi. Probably not oc'd and dx9 instead of dx8.
You are awfully upset that the 480 didn't live up to your expectations. If only there was a way to know how a GPU will perform in a certain game before buying it. Oh wait it's called a benchmark.

You made my day.
I should probably circle "Shadow of Mordor" in MSPaint but this'll have to do for now: http://i.imgur.com/DWJVGsV.png
Second link is 10/10, neither teh 780 Ti nor the 290X appear in those benchmarks. That means the 290X is faster, right?

Anticompetitives practices? nVidia crippling its own cards to make AMD look worse? See Hairworks. If you can turn it off, turn it off, if it's something you can't turn off let the lawsuits commence. It still doesn't make the 290X any faster, so while we're discussing performance it's not helping your case. I'd use Intel as an example why anticompetitive practices don't mean your product is worse, but you'd start defending your FX-8350 and the FX vs i7 debate has been dead for years, just like the FX series.

Next link I'm counting 2 wins for the 780 Ti, 1 for the 970, the 290X always being slower than either.
Project Cars pre-release 970 wins, 290X still slowest.
1.0.11 780 Ti is now close to 970, 290X barely beats normal 780.
1.0.13 and new drivers 2 for 780 Ti, 1 for 970, 290X still barely above 780, actually loses to 960 and 770 on 1080p.
Same GTA V benchmark as before.
Last GTA V benchmark: With the old drivers the 290X beats a reference 780 Ti on stock clocks. Now you've just got to find an aftermarket 780 Ti with less than a 7% overclock. Because once the boost clock touches 1000MHz that 3fps advantage is gone.

You just linked benchmarks that proved my point and show that nvidia is still updating the drivers for Kepler.

You don't even know which card I'm using. What if I told you I'm using a 7970?
But what is this smell? Could it be ...
http://www.teamfortress.tv/post/127945/how-well-do-amd-cpus-do-in-tf2

BUYER'S REMORSE?!
NOBODY SAW THAT ONE COMING!
Your 480 is Fermi, his 680 is Kepler though, I blame the 8350, but I can smell your salt from here.

Do I need to remind you that in Witcher 3 the 970 beats the 290X because of hairworks bullshit? If you take Witcher 3 with Hairworks as a benchmark both the 780 Ti and 290X are between the 960 and 970.

Link me all those "recent benchmarks". So far I've never seen the 290X win. Even against the 780 Ti on stock clocks. And I have yet to see a 780 Ti that failed to reach 1250MHz. There's also some 1500MHz 780 Tis out there that beat overclocked 980s quite easily but that wouldn't be a fair comparision.

Also define recent. I need to know wether I should dump 50 or just 10 benchmarks proving you wrong.

#9
You're the type of guy to buy a 480 or 290X and then complain about the temperatures, right?

Also how fucking delusional do you have to be to think a 970 beats a 780 Ti? Do you actually believe that somehow the 780 Ti became slower than the 290X?

I'm taking any bet that you'll buy an AMD card and come back whining about microstuttering and drivers again.

#1
980 Ti for best performance, 390 for best price to performance ratio. 390X and 980 are options if a 980 Ti is overkill, but a 390 doesn't quite cut it.

#3
The 285/380 is a thing.

#5
7950/7970 used probably mining cards.

I have never seen an nvidia card turn sour, but I haven't tried to drink any GPUs so far.

mid 2016 stretches the definition of "soon(ish)" a bit, don't you think?

#7
You know that lightboost works on both AMD and nividia?

In TF2 and CS:GO? Lots. I don't have any recent benchmarks for CS:GO and a TF2 update was just released, but the usual "300fps in 6s with maxfps cfg" shouldn't be a problem. It's also impossible to benchmark because fancy cosmetic items ruin fps, so depending on who you're playing against/with your fps will differ quite a lot.

Overwatch no idea.

It's fine, contrary to popular belief overclocking isn't free. It costs money and a lot of time/effort.

There's no reason to go 20% over budget if you don't have to. When radeon said "a wide range of games" I thought he meant AAA titles and for that the 390 made sense. For TF2 and CS:GO it's total overkill, like I said. If Overwatch is the only other game you'll be playing a 285=380 should be plenty.

Here we go, within budget, including monitor:
PCPartPicker part list / Price breakdown by merchant

CPU: Intel Core i5-6600 3.3GHz Quad-Core Processor ($229.99 @ Amazon)
Motherboard: ASRock B150M Pro4S Micro ATX LGA1151 Motherboard ($78.99 @ Newegg)
Memory: Crucial 8GB (2 x 4GB) DDR4-2133 Memory ($46.99 @ SuperBiiz)
Storage: Samsung 850 EVO-Series 250GB 2.5" Solid State Drive ($88.89 @ OutletPC)
Video Card: Sapphire Radeon R9 285 2GB Dual-X Video Card ($174.98 @ Newegg)
Case: Cooler Master N200 MicroATX Mid Tower Case ($43.99 @ SuperBiiz)
Power Supply: SeaSonic S12II 520W 80+ Bronze Certified ATX Power Supply ($44.00 @ Newegg)
Monitor: BenQ XL2411Z 144Hz 24.0" Monitor ($269.00 @ Amazon)
Total: $976.83
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2015-10-06 18:50 EDT-0400

For TF2 and CS:GO on 1024x768 the 390 is definitely overkill, I can't say anything about overwatch yet, but on low settings I doubt you'll need that much power to get 144fps.

For TF2 and CS:GO you should consider overclocking though. Would overclocking be an option for you?

What's your total budget, including the monitor? I didn't max out the builds so some money would be left for a monitor.

I'll adjust them (and narrow it down to one build) based on your answers.

PWM is easy, continous dimming is hard. It's not about lightboost, it's about normal operation.
Earlier VG248QE models used PWM at 360Hz to reduce the brightness, later models use 6*refresh rate (864Hz at 144Hz). The PWM is also synced to the refresh. Earlier XL24xxT models used PWM like the early VG248QE, later models and the Z series are PWM-free.

Most people don't experience any problems with PWM, only some are sensitive enough to get headaches from PWM above 200Hz and practically no one from frequency used on the newer VG248QE, but why take the chance?

There's also one thing that BenQ hasn't implented so far iirc.
Lightboost is very similar to PWM and since you want as little motion blur as possible the brightness will already be at the minimum you can tolerate, so you shouldn't need to lower the brightness. However let's say you're already at a fairly low lightboost setting but you want even lower brightness, without lowering the strobe length further. The strobes are 1.4 to 2.25ms. PWM technically wouldn't reduce the strobe length, since it turns the backlight of in the middle of the strobe. So what would you need to do to get 10% granularity, to reduce the brightness to 90% at the minimum strobe length. You have to turn of the backlight during 10% of the strobe, 1.4ms/10 = 0.14ms, so you need a backlight that's effectively capable of running at over 7000Hz. That's not happening. Simply using lower Voltage like the Zs do during normal operation is perfectly feasible though.