@RazorSharpFang@ENDESGA iirc the only major compiler that doesn't support it is msvc, though it's worth noting that each compiler that does implement this does so differently and has UB in odd cases (like passing 0).
@VallerieMomo@UltimaJyro Not an exploit: This is one of the core game mechanics, it is very intentional. The game teaches it to the player multiple times.
The sfx and yellow slash illustrate the intended cancel timing.
It doesn't seem to actually be measurably different, but it does look cleaner!
@RemithanVT That's fair, it's not really a consumer product. The reason to get one is if you want (potential for) better quality capture than the Elgato 4k Pro. Though it is worth noting that its cheaper than the other very high end capture cards while matching or exceeding their features.
@ReticentY2K I'm not Thiago Lira; and it's not right to call our work a "AI Decompilation", I'm pretty sure I have rejected every AI contribution that has appeared.
You're conflating the decomp with Thiago Lira's recomp.
@datnofact@theterk The CPU loses in single core but distinctly wins in multi-core. The issue here is the boost behavior and power limit on the steam machine. These two are very close, and in most gaming instances you're going to be in a multi-core workload, so it will be marginally better.
@catgirlprostate Survivorship bias. People who are willing to take the GPU out and put a new one in, are not going to visit a repair shop. They will work on their own and either tinker or RMA or take an L before they consider that. Plus, 8 years is historically enough for a platform upgrade.
@SMatweika@NateH8574@QuattroMKII Its a poor value and perf PC which is on a dead platform, whose core components cannot be changed, which without installing a different OS cannot play most big-budget MP games, the primary UX lacks core applications like streaming services. Console and PC downsides, no upsides
@dark1x@MKovyatkin There's Nuance. Winning is having better looking and performing games. Realtime lighting has benefits and drawbacks. Contention is that in most circumstances, its the worst option over baked and hybrid approaches for both visual fidelity and performance. Maybe it changes in time.
@_TAMA_GO @adilator @AerithClouds The "HD Textures" are the same PS2 textures xBRZ'd up 4x and had the contrast cranked. Specific characters were repainted (like Vaan's field model). This is unironically worse visually than the PS2 version.
@FluxedOver@TheBardess There's mathematically no reason to gamble on apex arrow because if you use it at 47.5 to burst at the latest it will always be 100 2 gcd before your 2 minute ends; so you just use it at 100, or at the nearest gcd to 47.5 at the latest to stay on track.
@TierZoo Continued: I would not suggest CPU encoding for this. It's also worth being familiar with the different encoding hardware that is in different GPU options; Nvidia's 70ti+ GPUs have an extra encoding unit, for example.
@TierZoo The client (viewer) decode is going to be pretty reasonable for most contemporary devices. The major issue you are going to run into is encoding. If gaming and streaming, you will likely need a second GPU or PC to have adequate encoding performance on GPU.
@Belcross_@TetoTetiroth It was the Master of Arms NPC, and in the final training, as well as the advanced guide, but they actually updated some of the text recently so it's a bit different than before.
@KanikouEstellia@BlueMightyDuck It is more ideal for a melee or prange to pull because they can provide damage reduction (additional arm's length) while doing damage sooner, and also improving mob grouping. With attention, you can also properly sprint chain. A clear can be up to 40% faster and much safer.
@bardozVAL@Blakeandmild The real answer to this question is that more counts/360 = less error, it's basic math. Though a lot of mouse sensors do interpolation nonsense or software scaling to support high and unusual CPI numbers; so you're typically at the mercy of your hardware anyway.