Windows 10 Gaming Performance Update

[AKK_K]

SLI บน DX12 ที่แรงกว่า SLI DX11
Deus Ex Mankind Divided 1080p High Preset DX12 SLI GTX960 2GB

FPS Average / Min / Max
DX12 SLI 63.8 / 48.3 / 84.2
DX12 SLI Factor ~1.76

DX11 SLI 52.3 / 41.5 / 70.4
DX12 Single 36.1/ 28.3 / 46.1 ~ DX11 Single

[AKK_K]

Deus Ex Mankind Divided 1080p Ultra Texture /All high /On GTX 960 2GB

[Shinnarat]

ขึ้นมาแบบแรงกว่าเดิมชัดๆ เลยแฮะ

[AKK_K]

VRAM Usage ของจริงหรือภาพลวงตา

Call Of Duty Infinite Warfare GTX 1080 Vs GTX 1070 Vs GTX 1060 Vs GTX 980 TI vs GTX 980

GTX 980 ใช้ VRAM MAX ~3.8-4GB
GTX 1060 ใช้ VRAM MAX ~5.9 GB
GTX 1070 1080 ใช้ VRAM MAX ถึง ~7.6 GB
และ
GTX 980@1265 VS 1060@1848 ใครแรงกว่ากัน

[AKK_K]

Using a GTX Titan X, which has 12 GB of VRAM, we tested the game's memory usage.
As you can see, the game uses a ton of video memory, which is in line with what we've seen from previous Call of Duty titles.
However, as the performance results in the next section show, even cards with less VRAM have no problems delivering a good framerate,
so it's probably just the engine trying to fill up video memory as much as possible, even with assets that might not be needed at all.

https://www.techpowerup.com/reviews/Performance_Analysis/Call_of_Duty_Infinite_Warfare/4.html

Conclusion

VRAM usage of Call of Duty Infinite Warfare is a curiosity.
In the past, we've seen excessive video memory usage from Call of Duty and Infinite Warfare is no exception.
Despite showing video RAM usage of up to 8 GB,
cards like the GTX 1060 3 GB or RX 470 4 GB are delivering framerates that fall in line with their expected shading performance.
This is definitely good news for people who - for economical reasons - bought cards with smaller memory amounts.

https://www.techpowerup.com/reviews/Performance_Analysis/Call_of_Duty_Infinite_Warfare/5.html

[Gintama023]

กระแส 2GB ปรับสุดยันปีหน้ากำลังจะกลับมา

ถ้าแรม 2GB มันพอจริงป่านนี้คงมีแต่คนใช้ไปแล้ว คงไม่มาเสียตังกะพวกแรมเยอะๆ แพงๆ หรอก

[AKK_K]

Deus Ex Mankind Divided ในส่วน Texture ปรับแต่งได้ 5 ระดับ
คือ Low / Medium / High / Very High / Ultra
ตั้งแต่ Very High ตัวเกมแนะนำว่าควรมี VRAM มากกว่า 4GB

แต่เกมมีการ Update มาจนถึงจุดที่เรียกว่าดีพอสมควร
และประสิทธิภาพของ Nvidia ดีขึ้นกว่า day 1 ~ 30-40%
มาดูผลการทดสอบที่ Very High Texture

Deus Ex Mankind Divided 1080p Very High Texture / All High/On GTX 960 2GB Single

[AKK_K]

AC Syndicate

Max Texture (ต้องการ 4GB VRAM )
PCSS Ultra
HBAO +
TXAA 4x + FXAA

บน GTX 960 2GB

[AKK_K]

Rise of the Tomb Raider 1080p Very High Texture 3.6 GB Max Vram Consume On GTX 960 2GB

Very High Texture คือ HD Texture ที่ตัวเกมบอกว่าต้องการ VRAM > 4.0 GB

[AKK_K]

Techno-kitchen confirm
4.8GB VRAM Consume ON GTX 1060 3GB

Watch Dogs 2: GTX 1060 3 vs RX 480 8 (Ultra)

4801/3035 MB แสดงในหน้า set up ของ GTX 1060 3GB

[AKK_K]

5.1GHz I7-7700K Gaming/3D
http://www.bit-tech.net/hardware/cpu...r /> preview/2

[AKK_K]

CPU Update

The "Zen" Core Architecture

ี่นี่ก็เป็น 1 core นะครับ
1 core จะมี 4 Interger Unit + 2 Floating Point Unit และที่สำคัญ 2 Thread / Core

Here's Zen's high-level core diagram. There are several significant differences compared with AMD's older Bulldozer core,
including the addition of an op cache, a micro-op queue, and a larger number of integer pipelines per core.



Execution, Load/Store, INT and FP Scheduling
The execution of micro-ops get filters into the Integer (INT) and Floating Point (FP) parts of the core,
which each have different pipes and execution ports.
First up is the Integer pipe which affords a 168-entry register file which forwards into four arithmetic logic units and two address generation units.
This allows the core to schedule six micro-ops/cycle, and each execution port has its own 14-entry schedule queue.

[AKK_K]

2 Thread / Core

AMD moved to an SMT (Simultaneous Multi-Threading) design with Zen,
as opposed to its old clustered multithreading technique that aimed to keep all of the structures fully available to increase single-threaded performance.



Simultaneous MultiThreading (SMT)

[AKK_K]

ZEN CPU Complex
Where Bulldozer used the concept of a CPU module (defined as a pair of cores that shared resources), Zen uses complexes.

AMD arranged the CPU complex (CCX) into four cores connected to a centralized 16-way associative 8MB L3 cache
that it split into four slices. All of the cores can access the entire pool of L3 cache (purportedly with the same average latency).

[AKK_K]

ZEN Cache architecture
AMD has stated that L1 and L2 bandwidth is nearly 2x Excavator while L3 bandwidth is supposedly 5x higher.
These changes should keep the core fed and support higher performance.
The L1 cache is write-back instead of write-through that's a significant change that should improve performance
and reduce cache contention
(Bulldozer?s write-through cache meant that L1 performance could be constrained by L2 cache write speed in some cases).