Yeah. 1TB or more?
Probably wouldn't overclock if you're going for such a small form factor.
Keep in mind that at that size everything's a bit tight an the build process is a bit more difficult. I generally recommend µATX because full ATX is usually a waste of space, but mini-ITX is possible. Just be aware that this will massively limit your choice of mainboards and coolers and potentially limit your choice of GPUs. The mainboards will be more expensive and the coolers of that size will obviously be limited in what they can do so overclocking can be fairly limited up to almost impossible.
Maybe look up the sizes of small µATX cases that are still easy enough to build in (20-30L volume) and compare that with the 10-15L category of mini-ITX that you're looking at right now. You don't gain that much so unless space is really tight I wouldn't recommend it.

Technically yes, practically no. M.2 has basically replaced mSATA so you'll need an adapter but it shouldn't change the rest of the build. Which model is it anyway? Maybe if the budget allows for it you could get an upgrade anyway.

You don't need an HDD either?
Linux should be compatible with more or less all desktop parts, so you don't need to worry about that.
Stellaris will run like dogshit in the endgame on any hardware so you don't need to worry about that either.

I'm guessing you'll want to build this asap?
Only one question left, overclocking yes/no?

Not worth it like others have said, it would be a minimal difference.

Single threaded performance although passmark is not the most accurate for that.
But a 2600X at 4.0 will run circles around an 870K at 4.0 Ghz.

The tradeoff is more cores + lower clockrate vs fewer cores + higher clockrate, not Intel vs AMD.

Yeah, proper upgrade for 400-ish would be better and last longer. Maybe less. Even a 2200G would be an upgrade and we don't know what the prices for the next gen will be like.
In TF2 the 1060 is definitely not a bottleneck. What games are we talking about?

Not really. AMD CPUs are better on a budget right now because they're simply cheaper for a given core-/threadcount and all of them are overclockable. E.g. 4 cores/4 threads starts at 99$ and 3.7 GHz for AMD and you could still oc that, Intel starts at 117$ for 3.6 GHz (roughly the same single threaded performance) but it's locked. At 6c/12t it's 199$, 3.9 GHz overclockable vs 303$ 4.6 GHz locked. Sure the 4.6 GHz are nice for games but if you're on a budget it's not really a choice. If you need the threads it's obvious what you should buy, it's just that if you don't need the threads then there are some reasonable Intel options available like 6c/6t 4.1 GHz for 182$. Both are perfectly reasonable good all-around sub-200$ CPUs, but one of them got a massive advantage if you use more than 6 threads.

tl;dr
Wait, save up some money, see what Ryzen 3xxx brings. Maybe you don't need a 6 core CPU. Maybe 4 cores/8 threads is perfectly fine. If there's a Ryzen 3xxx CPU with 4 cores and high clockrates for 150$ then that's a better choice for games than the 2600X.

Your summoning power is weak af. Novices should at least draw a proper summoning circle to act as a multiplier.

Ok, I'm going to ask: How many fps were you expecting?

The FX-8800P is an unimpressive laptop APU and performs like an unimpressive laptop APU. The R9 360 isn't exactly high end either.

If you can turn off the integrated GPU (you don't need it, only steals from the power budget for the CPU), crank up the TDP/power limit as high as it can go and overclock the CPU.

I'm guessing your current mobo doesn't support M.2 SSDs and it wouldn't be that much of an upgrade anyway so just keep your current SSD.
EDIT: If you really want an upgrade though and got enough money left over (depends on the GPU) you can look for one of the cheaper PCIe SSDs.

Overclock if you haven't already. Overclock higher if you already have. :D

Buy a new GPU. Considering your budget you can get anything, it's just a matter of how many fps you think you'll need/want. E.g. if you want twice the fps or a bit more then get a 1080 Ti or 2080.

#3220
That's a decent SSD so NVMe would be an upgrade but not game changin like HDD->SSD.

Apex is a bit difficult to benchmark (much lower fps in MP than in the benchmark), but that still seems a bit low.
Are you sure that nothing is throttling?

Either way I don't think that the CPU is the problem here.
Probably the most comprehensive and realistic benchmark would be this:
https://www.computerbase.de/2019-04/apex-legends-leserbenchmarks/
These are max settings but you can see that it's pretty much sorted exclusively by GPU performance.
Most interesting results in your case:
GTX 1060 CD, R5 2600, W10
69,4
GTX 1060 FE, i9-9900K, W10
64,0

and

RX 580 CD, E3-1240v2, W10
80,5
RX 580 CD, R5 2600 OC, W10
79,0
RX 580 CD, i9-9900K, W10
76,5
RX 480 CD, R5 1600X, W10
76,0

In both cases the slowes CPU gets the highest fps because it's paired with the fastest GPU.
Aftermarket 1060 > Founder's Edition 1060.
580 only slightly faster than a 480 with the 9900K doing fuck all.
Different 580s with different clockrates getting different results and a 1240v2 (slower than the 4670K apart from Hyperthreading) easily beating a 9900K.

Apart from higher minimum fps more threads don't see to do that much either so I really don't think that's the problem.
https://www.dsogaming.com/wp-content/uploads/2019/02/Apex-Legends-CPU.png
That's with a 4930K, faster than the 1240v2 but slower than the 4670K assuming you overclocked it even slightly, except for the core/thread count.
So with the exception of minimum fps the 4670K should not be slower than a 1240v2 (or 4930K with 4C/8T) enabled and those are GPU bound at max settings 1080p with a 1060/580 and should easily get over 100 fps on 720p.

EDIT:
After looking at some 720p/1080p and High/Low benchmarks scaling seems to be dogshit, going from 1080p High to 720p Low doesn't get you much more fps.
So the computerbase benchmark is fairly accurate and 60-65 fps on 1080p High translate to the 90-ish fps you're getting on 720p Low.

Conclusion: It's the GPU.

#3221
If the 4670K is seriously underpowered then what CPU isn't?

Not sure if there are SSDs that can't do something.

#3222
Well what's the build going to be used for and when are you going to buy the parts?
New CPUs this summer and all that.

Then maybe an NVMe SSD could make sense. Current SSD?

Also what's your current CPU? Unless it's pretty bad I don't see how upgrading it would improve fps in new games.
Any examples of what games you'd be playing and how many fps on what settings you're getting and how many you want?

Intel socket size hasn't changed in the past 5 years so the cooler will still fit.

What's the goal of the upgrade?

A good NVMe SSD that's actually fast will cost you so you'll have to give up other things (do you really need 4000 MHz RAM?) if you want to make it happen within that budget. If you're not doing anything with a shit ton of random reads/writes the difference between a good SATA SSD and a reasonably priced NVMe SSD isn't nearly as noticeable as the upgrade from an HDD to an SSD.

We'll deal with the mobo later. If you want to overclock you'll need a good cooler. If you don't then a lot of this build should be changed.

Ok, assuming you don't want to wait until summer for new CPUs there are a few options.

No overclocking would mean something like this with a CPU between an i3-8100 and an i7-8700, depending on how much you're willing to spend.
https://fr.pcpartpicker.com/list/6XmFV6
https://fr.pcpartpicker.com/list/ZBfFV6
Keep in mind that it's mostly about clockrate so the difference won't even be that much. Even the 8100 would be enough for your purposes.
At that point going with an AMD CPU with higher clockrate and faster RAM will probably work out to around the same or better performance at a lower price.

Now if you want faster RAM because it does help in TF2 you need a Z370/Z390 mobo and at that point adding a cheap cooler and switching to a 8350K would allow you to overclock. Or go all the way with a decent mobo and cooler and an i5-9600K. Way overkill but obviously significantly better and still doable with your budget. https://fr.pcpartpicker.com/list/fQ3mtg

So think about it, if you don't want to invest the time needed for overclocking a ~650€ build will do what you want it to do just fine. I mean if your old laptop was fast enough this is guaranteed to be more than fast enough.
Basically what I'm saying is it's not worth paying for the option of overclocking and a few percent more performance if you end up not doing it and would be content with less so I'm asking you if you'd also open to spending less than 900€ even though it means you're not getting the latest and greatest, the best upgrade possible within your budget. I mean you can buy a FreeSync monitor or whatever you want with the leftover money.

HDD can be added if you need more than 500GB, case depends on your preferences more than on anything else. I went with µATX because why not. No need for a bigger case.

#3210
Overclocking yes/no?
When are you going to build it?

#3212
I'm guessing that's a no for overclocking?
Can you be a bit more specific with the "other shooters"? Getting 60 fps on low in CS:GO is quite different from getting 120 fps on high in Quake Champions even though neither is particularly intense graphically.

Why would you blame AMD for Shadowplay fuckups?

Setting lerp to 0 causes hitboxes to be inaccurate.
That has been public knowledge since around 2007-ish.

Cmdrate 100 is useless on 66 tick servers.
Updaterate is limited by the server by simply not sending you more than 66 updates per second even if you request 10 million. Interp is still calculated with interp_ratio/updaterate without the limit applied because of reasons.
Again, this has been publich knowledge since around the time when cl_interp_ratio was introduced.

I rate 2/10.

It's confirmation bias.
Your aim just sucks after 1h and then you look for an update to blame.

Dr. Setsul prescribes one dose of

https://i.imgur.com/NqpPXHu.jpg

Antivirus isn't hardware so do I really care?

Even worse this is obviously a psychological issue. The lingering doubt in your mind that your pc is not as secure as it could be takes away subconscious focus from your aim.

Correct, but can't you see the problem?
The goal isn't to have the team with the most wins or the least losses win. The goal is to have the best team win.

So if those teams are 1-1 against each other which one is better? One team won the first match, then lost in the winners bracket where arguably the stronger teams should be. The other team lost the first match then won against weaker opponents in the lower bracket and then won the second match. It's not obvious which team is actually better.
That's why using longer matches for later rounds makes sense. If you lose a Bo3, but win a Bo5 it would make more sense if you're the better team, statistically speaking. Especially if you win the Bo5 against higher seed advantage due to coming from the losers bracket.

By the same logic using a match of around twice the length for the final and no bracket reset also makes sense.

The basis for best-of matches is that the team that wins more maps is better. If you play a Bo3 of Bo3s you can win with less maps won. That's a direct contradiction to the assumption the format is based on.

So two matches for the grand final is obviously not the only fair choice because it isn't even fair. Playing a Bo3 in the upper bracket finals and a Bo7 in the grand finals is obviously fair from the perspective that the better team wins. But what advantage do you give the team that won the upper bracket final? A Bo5 also allows winning with a drawn map score. Bo4 is fairly reasonable in terms of overal map score but Stark complained about the "random map advantage". So no, the choice is not obvious.

Usually with long game times where everything else is Bo1 and Bo3 isn't feasible or where every game will be played on a different day playing two matches with a bracket reset makes more sense, but that's not the case here.