AMD Vega - Info, spekulace

[del42sa]

http://www.digitimes.com/news/a20170824PD207.html

Some sources pointed out that Vega's design of integrating high bandwidth memory (HBM) into the GPU has significantly raised the difficulty of the related packaging process and hence resulted in low yield rates. However, some other sources claimed that the issue may be down to problems with Advanced Semiconductor Engineering's (ASE) packaging technology.

[spajdr]

unwinder ohledně podpory Vegy v Afterburneru.:
- My Vega 64 sample just arrived so I'm starting implementing support for it. Fan control will be the most tricky part: unlike the previous cards Vegas do not support fan control via duty cycle adjustment (i.e.fan speed adjustment in %), do not support setting desired fixed fan speed directly, so it will be rather troublesome to provide software fan adjustment curve for it.

[webwalker]

Video z GPUlab:
Frekvence, limit spotřeby, napětí, undervolting

To je AMD Radeon RX VEGA 64, seznamte se prosím! >>

vega-video.jpg

[/url]
Omluvte zhoršenou kvalitu zvuku. Ani třetí výměna mikrofonu to nespravila

[spajdr]

Kvalita zvuku je v pohodě, pokud nerveš mikrofon hned pod větrák grafiky.
Jinak díky za test

[del42sa]

AMD loses $100+ on every Vega

[Hladis]

Ten clanek mi nedava zadnej smysl.

but Vega for AMD is not about making money. Don’t get me wrong, every company would like to make money with every product that it makes, but for AMD it is more important to win market share. First you win the market share, then you go after better ASPs (Average Selling Prices) and potentially start running a positive business.

Tohle co tam napsal je totalni nonsense. Vega není karta, ktera by mela "win the market share" a pokud by na tom ještě "udajne" prodelavali, tak to je proste fail jak swine.

[del42sa]

Podle me je to blbe napsane, urcite na tom nevydelavaji (nizke az nulove marze) ale urcite na tom neprodelavaji, teda pokud nejsou uplne mimo. Vyrobcum AMD marzi vzit nemuze, protoze jinak by se jim na tu vyrobu vyprdli, ale AMD v ramci ziskani podilu nemusi mit marzi zadnou nebo jen symbolickou. Urcite ale neprodavaji stylem, ze by kazdou kartu dotovali 100$ a pokud to delaji, tak jsou fakt zrali na bohnice.....

Navic tohle se tyka jen hernich karet, profi karty urcite neprodelavaji.

[Hladis]

To je jasny, ze u profi by to vydělat mělo kvůli vyssim marzim a cenam. Ale neverim v to, ze by dotovali každý kus herni vegy 100 dolary, to by bylo fakt na zavreni Koduriho k Chocholouskovi. To se vzdusnym zamkem, ze by jim to hodne nakoplo market share a tedy prodělek není dulezity, to by nebylo k Chocholouskovi ale rovnou na utraceni. Podle me je ten clanek nejak zmotanej.

[icebosss]

však někde snad byly čísla s cenou gpu, hbm, pcb atd ne? Tam cena vycházela za vegu cca o 20-30 dolarů na výrobu víc než za hawaii nebo s čím tam bylo to srovnání.

[Hladis]

Nejaka cisla tu kolovala, ale to bylo k Fiji.

[Hladis]

Koduri hodil davu na twitter něco o Veze http://wccftech.com/amds-raja-koduri-br ... -features/ U "Vega10 has largest perf/watt dynamic range than any GPU I recall" jsem se zakuckal.


//edit Wilik - promaz OT

[Krteq]

Nějaké info od sebbiho

Vega has some advantages, but none of them are seen in current games or current software. It's hard to predict how much these new features will matter in the close future (when Vega is relevant).

For gaming, the biggest change in Vega is full DirectX feature level 12_1 support. Both Intel and Nvidia had already released two GPU generations with these features. These features allow implementing more efficient algorithms for many rendering problems. Now that AMD also supports these features, we can expect more game devs to start adding support to their games. AMD has slightly higher DX12 feature level support than Nvidia regarding to some aspects. Resource heap tier 2 (vs 1), conservative raster tier 3 (vs 2) and support for stencil ref from PS. My prediction is however that none of these slight advantages will have an actual impact during the Vega life time. But the common 12_1 features (that are supported by all IHVs) will certainly be used in some games, and that's when Vega will prove to be highly superior compared to previous AMD cards. There are also some other new gaming related features (such as primitive shaders) that are not yet exposed to developers. Whether these features will be used during Vega life time (on PC) remain to be seen. Evolution of PC graphics APIs hasn't been that fast lately. I don't expect big changes that only benefit one IHV. Consoles could of course be a good proving ground, but it will take some time until Vega base GPUs reach consoles.

For compute, Vega has double rate fp16 support. This is handy for some compute problems, such as deep learning. Nvidia's consumer cards don't have this feature (it's limited to professional P100 and V100 + mobile). Vega also has a brand new CPU<->GPU memory paging system. The GPU on-demand loads data from DDR4 to HBM2 at fine granularity, when a page is accessed. This allows 8 GB Vega to behave like a GPU with much larger memory pool, assuming that the memory accesses are sparse. This is a common assumption in games and also in many professional software. Usually games only access a tiny fraction of their total GPU memory per frame and the accessed data changes slowly (smooth camera movement and smooth animation). This could be a big game changer if games start to require more than 8 GB of memory soon. If a 8 GB Vega behaves like a 16 GB card with this tech, it could greatly extend the life time of the card. The same is of course true for many compute problems. However Nvidia Pascal also has a similar GPU virtual memory system for CUDA applications, but their tech doesn't work for games. CUDA has an additional page hint API to improve the efficiency of the automated paging system. Developer can tell the system to preload data that is going to be accessed soon. Whether or not the automated paging system will be relevant during Vega life time depends on software and games you are running. Xbox One X console with 12 GB memory (and a 24 GB devkit) is coming out very soon. Maybe we need more than 8 GB on PC soon. Let's get back to this topic then.

Currently the appeal of Vega is mostly forward looking. There's lots of new tech. AMD was lagging behind the competition before, but now they have a GPU that has lots of new goodies, but no software is using them yet. It remains to be seen whether AMDs low PC market share is a roadblock for developers to adapt these techniques, or whether AMDs presence in the consoles helps AMD to get high enough adoption for their new features.

Beyond3D forums

[Hladis]

Občas trochu nadsadil, ale udrzel se u země, ze to hitparáda nebude. Tech par featurek, který zni dobře tak na papire, není zadny velky deal, který by z Vegy mohl vytahnout kvanta vykonu navíc v globale a sam je skepticky, ze se to vůbec vyuzije za života Vegy. Vega se musí spoléhat ciste na milost a nemilosat vyvojare, aby nejaky featury vyuzili a trochu ji pomohli. Veze paradoxne muze "pomoci" vydani Volty, která muze prinyst další zmeny smerem k DX12 a compute, takze z toho muze tezit i Vega, akorát to bude uz jaksi rok po dvanacty. Jsem dost velkej skeptik k tomu prezentovanymu, ze by 8GB VRAM zaclo byt rychle bottleneckem.

[maltra]

Pre mna je problem vytesovat sa z featur ktore karta sice ma ale nie su vyuzivane prave teraz, teda hned po vytiahnuti z boxu. GPU vyvoj je rychly a za rok uz to nebude nikoho zaujimat aka rychla je Vega s podporovanymi funkciami, pricom ta software podpora tiez nie je 100% zarucena. Musim ale povedat ze na to aky pruser ta Vega hned po uvedeni bola, AMD to docela ustalo a nejake zemetrasenie nenastalo. Stale sa o Vege hovori a stale sa predava takze snad nejake peniaze na vyvoj z toho mat budu.

[Hladis]

No Vega je pruser, ale diky stavu trhu i takovy pruser, který by v jiny konstalaci skončil fakt spatne, se ted slusne proda. Minerum je vcelku u zadku herni vykon. U herni verze se spekuluje, ze z ni AMD moc vydelek nemá (viz. vyse), spis budou sazet na profi, kde by to mohlo něco trhnout.

[icebosss]

Zas takovy pruser to neni kdyby tu byly normalni ceny. Ja si nemuzu stezovat na vegu 56 za 11600, porad slusna cena s dobrym vykonem po undervoltu. Uz se tesim az prijde, snad tenhle tyden. Porad to bude boost oproti furyne na qhd u novych her

[maltra]

Co ja viem, az taky boost to proti Fury podla mna nie je. Aspon co som pozeral prve testy tak sa mi to nezdal velky upgrade. Ale zalezi na ake hry to chces, ja som zamerany na Witchera 3 hlavne.

[icebosss]

já prostě na nový hry a už mi dost vadí ty 4gb vram na QHD, ta vega na 1700mhz si myslím bude slušný boost oproti fury, kor s fullcover blokem nebudu rešit hluk a teplo

[del42sa]

zdá se že DBSR je stále neaktivní, včetně ostatních nových GFX9 featur...

https://www.phoronix.com/forums/forum/l ... post970697

Draw Stream Binning Rasterizer is not being used yet in the open drivers yet. Enabling it would be mostly in the amdgpu kernel driver but optimizing performance with it would be mostly in radeonsi and game engines.

HBCC is not fully enabled yet, although we are using some of the foundation features like 4-level page tables and variable page size support (mixing 2MB and 4KB pages). On Linux we are looking at HBCC more for compute than for graphics, so SW implementation and exposed behaviour would be quite different from Windows where the focus is more on graphics. Most of the work for Linux would be in the amdgpu kernel driver.

Primitive Shader support - IIRC this is part of a larger NGG feature (next generation geometry). There has been some initial work done for primitive shader support IIRC but don't know if anything has been enabled yet. I believe the work would mostly be in radeonsi but haven't looked closely.

For both DSBR and NGG/PS I expect we will follow the Windows team's efforts, while I expect HBCC on Linux will get worked on independently of Windows efforts.

[Dolan]

AMD sa na 7nm pravdepodobne vráti k TSMC.

http://www.digitimes.com/news/a20170908PD210.html