Gaming Tech Talk

A koliko takva konfiguracija košta ?
Ne znam vam ja niš o PC-u pa eto prosvjetlite njubaru.
U svakom slučaju mislim da nije baš jeftina.

S autom nemozes gejmat  8)

Bez kucista, napajanja, monitora i HDDa cijena ovih komponenti je cca:

Geforce GTX 970 Overclocked
Geforce GTX 660 Physics only
Intel I7-5820K
8,00 GB Ram DDR4
Asrock X99M Extreme4
-------------------------------------
11000 kn

Iskreno totalno neisplativo danas, jer mu je dosta otislo na nepotrebni ddr4 i taj cpu.

Blizu sam bio, računao sam i monitor 

https://youtu.be/JxUPJdcChzE

Sve cross gen igre...baš me zanima kako bi se ta ista konfiguracija nosila sa current gen only igrom.
Anyway dual core nije ni malo future proof tako da je totalno neisplativ,osim ako netko želi igrati samo last gen igre na pc-u.

Ja sam dugo vagao sto uzet za svoj budget build i na kraju se odlucio bas za ovu kombu iz videa(g3258+msi h81-p33)..al nemam jos graficku..mislim da je procesor cisto ok za 500 kn kolko me kostao prije ovog ogromnog skoka dolara,za igre kakve preferiram bit ce mi dovoljan..ionako mi je gaming nekako pao u drugi plan u zadnje vrijeme

https://youtu.be/QhlLypEJq-c

U većini slučajeva stigne update koji popravi ovakve stvari tako da thumbs up za to iako devovi ne bi trebali fušati.

Pošteno od njih da su izbacili update,ali dalje ne umanjuje sranje koje su prvotno izbacili. Očito je takva situacija sa igrama,da je bolje što brže izbaciti igru pa je poslije patchati nego malo pomaknuti deadline pa izbaciti kompletan proizvod.

AMD explained today two major features that will be included in Microsoft’s new API, DirectX 12. These two features are Async Shaders and Multi-Threaded Command Buffer Recording. DX12 promises to give extra life to all available older multi-core CPUs, and these two features will be essential to that task.As AMD claimed, Async Shaders is a feature allows a game engine to execute GPU compute or memory activities during “gaps” in the graphics workload presented by a game.    “In DirectX 12 Async Shaders supercharge work completion in a compatible AMD Radeon GPU by interleaving these tasks across multiple threads to shorten overall render time. Async Shaders are materially important to a PC gamer’s experience because shorter rendering times reduce graphics pipeline latency, and lower latency equals greater performance. “Performance” can mean higher framerates in gameplay and better responsiveness in VR environments. Further, finer levels of granularity in breaking up the workload can yield even greater reductions in work time.”AMD then talked about Multi-Threaded Command Buffer Recording, and explained the great benefits that this feature will bring to all DX12 games.In DX12, the command buffer behavior is radically overhauled in five key ways:        Overhead is significantly reduced by moving driver and API code to any available CPU thread        The absolute time required to complete complex CPU tasks is notably reduced        Game workloads can be meaningfully distributed across >4 CPU cores        New “bandwidth” on the CPU allows for higher peak draw calls, enabling more detailed and immersive game worlds        All available CPU cores may now “talk” to the graphics card simultaneouslyOxide’s Ashes of the Singularity, an upcoming strategy game that already utilizes all 8 cores of an AMD FX-8370 processor, will be one of the first games showcasing the benefits of Multi-Threaded Command Buffer Recording.As AMD and Oxide noted, Ashes of the Singularity offers image quality and resolutions that—in the words of the developer’s CEO Brad Wardell—are “not even a possibility” under DirectX 11.http://www.dsogaming.com/news/amd-explains-two-major-features-of-dx12-async-shaders-multi-threaded-command-buffer-recording/

Šta mislite o ovom laptopu. Kontam da ga mijenjam za X1.
http://www.olx.ba/artikal/16949384/alienware-laptop/

https://youtu.be/u7QO6VstDfM

Many tech demos end in tears. Square Enix’s moving demo during the keynote at Microsoft’s BUILD 2015 developer conference this week might be the first where that’s exactly the point.

In collaboration with NVIDIA and Microsoft, Square Enix – maker of the Final Fantasy series of video games – stunned the BUILD audience with a research project called WITCH CHAPTER 0 .

The project portrays human crying — considered one of the most difficult emotions to recreate digitally — at a level of quality never seen before with a real-time, computer-generated character.

It’s a demo that may come to represent a milestone in an industry that has long-struggled to cross the “uncanny valley.”
- See more at: http://blogs.nvidia.com/blog/2015/04/30/build-demo/#sthash.1bGyDu7i.dpuf






Coined by pioneering roboticist Masahiro Mori nearly 50 years ago, the term describes the boundary where emotional realism in the digital world begins to blur with reality — and creates a sense of the strange, or even revulsion, in human observers.

Overcoming this sensation has been a stretch goal for computer-generated graphics for decades. It’s a task that’s even tougher when performed in real time.

WITCH CHAPTER 0 project, powered by Microsoft’s new DirectX 12 application programming interface (API) and NVIDIA GeForce graphics, points to a way through the uncanny valley.

Bringing more reality and depth to character expressions will better immerse players in stories during game-play, and deepen their connection to characters.

NVIDIA GameWorks Effects Studio and GeForce GTX graphics were among the next-generation technologies Square Enix put to use. Square Enix also conducted extensive research on real-time CG technology using DirectX 12. The results will be incorporated into Square Enix’s Luminous Studio engine.











During the BUILD conference, Microsoft unveiled the research project running on four of our flagship NVIDIA GeForce TITAN X GPUs. It’s a stunning example of what they can do with the DirectX 12 API, due to arrive with Windows 10, Microsoft’s next-gen operating system.

http://channel9.msdn.com/?CR_CC=200605262

Much of the know-how achieved by this project will surely trickle down to tomorrow’s games. That’s reason enough for tears of joy.



https://youtu.be/e1mwhdYZvSE

http://blogs.nvidia.com/blog/2015/04/30/build-demo/

gledajuci po tome sto bi dx12 mogao donjeti uz HBM2 kartice koje dolaze 2016/2017 godine mogli bi vidjet velike pomake u svemu

6-8 jezgri simultanog rada kako spada
graficke sa 8-16gb memorije i naspram sada x4 na vise propusnost i opcenito brzine

ako ne veliki pomak po pitanju renderanja odredenih stvari onda veliki pomak u rezoluciji i FPS-u za one koji zele 120hz il 144hz monitore u 4k

Yap, a samim time razni filteri i samplinzi gube smisao na takvim rezolucijama, te se stvarno može posvetiti i fizici modela i okoliša s obzirom najavljeni broj aktivnih jezgri.

cak i ne samo to

evo jedno pitanjce
ako igra ima 1000npc-ova na mapi
jeli teze renderat 1000 klonova ili da je svaki jedinstven?

veliki post u spoileru
napomena brojke nisu egzaktne nego reklo bi se matematika na salveti samo da sluzi poanti, kao i previse pojednostavnjeni primjeri makar tako doslovno nemora bit

sada ako ste se pitali i dali neki odgovor da pojasnim
1000npceva na mapi kada se rendera mora se renderat odreden broj poligona, svaki lik ako su iste kvalitete nebitno jesu li klonovi ili jedinstveni ima isti broj poligona
kvaliteta modela ovisi o tome
sada kakve to ima veze ako kazemo da svaki lik ima 10k poligona to je 10 000 poligona x1000npc
tako da je prakticki jednako jesu li svi jedinstveni ili ne
ali zasto sada nisu svi jedinstveni kada je tako "lako"
nisam rekao lako, rekao sam da nije bitno po pitanju poligona
kako bi bili svi jedinstveni potrebno je 1000 razlicitih textura, bilo to face il obleka
pa kako to otezava? zahtjeva jako puno memorije i ta memorija se mora moc brzo punit/praznit itd.. znaci brz rad s istom

opet kako i zasto je to bitno?
pa sto rade HBM kartice i dx12 sam po sebi?
HBM kartice udvostrucuju danjasnju memoriju i ucetverostrucuju tu memoriju od sljedece godine
isto te kartice imaju puno vecu, sada 9x propusnost naspram danasnjih a primjenjivo su sa 5x nizim memory clockom 2x brze
nisu egzaktni brojevi samo da se zna
opet kaj onda?

pa puno vise memorije, puno brzi dohvat te memorije, brze se moze napunit
analogija:
imate kadu punu vode i zastopana vam cijev u kadi a svu vodu cete presipat u wc
imate zlicicu ili kanticu
kantica je brza i to puno jer u nju odjednom stane vise vode

znaci memorije se moze nabit lako ali ako ju se nemoze iskoristit kako treba ona je tamo za bezveze
sto se brze to moze koristit a dx12 smanjuju vrijeme cekanja i samo nepotrebno smece da se optimizirava citav proces rada cak do 30tak posto koliko sam procitao
uz HBM kartice koje mogu zaprimit vecu kolicinu memorije a posto su sve jace i jace naravno odradit sve to

zalosno je da danas previse performansi otpada na idle period ili na to sto kartica nemoze primit informacije koje je ipak sposobna odradit makar imala 4gb nije dobila od toga nista

zato recimo ako kartica ima bandwidth 300gb/s sada ce imat 600 i preko toga
od sljedece godine zele imat minimum 1tb/s

a to ce bit puno veci pomak nego u zadnjih par godina uz dx12 vise jezgri ce bit benefit a ne porez na budale sad


4K rezolucije ce ubit potrebu za filterima, vise jezgri i bolje iskoristavanje ce omogucit napredku u polju fizike, AI i efekata
jace graficke kartice sa brzom memorijom i puno vise isto ce omogucit kompleksnije scene pogotovo u dx12 koji ima 10x vise draw calls, puno bolje texture ogromnih rezolucija i puno vise textura po raznim slojevima, znaci ogrebotine metala, unistavanje robe na likovima, trosenje boje na povrsinama itd..

kao ovaj squeenix video
reklo bi se big deal?
koza je masna sama po sebi, voda ima svoja opticka svojstva a koza ima isto reflektivna svojstva po slojevima
kako bi se postigla realna "simulacija" svega toga, kako se voda krece po kozi i kako nesto upija i reflektira svijetlost nije jednostavno tako da samo u tome je video impresivan

tako nesto necemo vidit uskoro ali ako onakav napredak stize vec 2016 godine bilo bi bedasto ocekivat da za jedno 4 godine sa 2 generacije novijim hardware-om necemo moc tako nesto makar ne apsolutno real time i imat

https://youtu.be/pU70-e0XHxw

Oni su kompletno igrali bez sharpeninga i AA!