AMD CEO : Vega & Customer Relationship Key To Winning 50% Share
Yesterday, AMD’s CEO Dr. Lisa Su shed light on the company’s ambitious plans in the graphics market and its on-going efforts to set them in motion. Speaking at the 20th Credit Suisse Annual Technology, Media & Telecom conference Dr. Su underlined the key to growth and market share expansion in the discrete graphics market.
2016 has been a very busy year for AMD, highlighted by the launch of its next generation Radeon graphics products based on the brand new 14nm Polaris architecture. The company’s focus on affordable Radeon RX 400 series gaming products paid dividends and it saw its market share nearly double over a 12 month period after it had lost much of it in the year prior.
However, it saw its graphics share momentum slow down in the third quarter, as a result of new product launches in the entry-level market segment from its rival Nvidia. It remains to be seen how things will play out in Q4 of 2016. Looking forward, AMD is poised to re-ignite the engine of momentum next year as it expands its graphics product line with a new Radeon family based on the high-end Vega graphics architecture.
AMD’s Vega GPU Family Officially Launching In The First Half Of 2017 – To Deliver Enthusiast Class VR & 4K Performance & Features
According to everything we’ve been hearing so far, Vega represents a significant architectural leap in performance and power efficiency. The Vega 10 GPU is an especially impressive one. Purported to deliver more than double the graphics horsepower of the RX 480, double the memory and double the memory bandwidth. It’s also officially AMD’s first ever graphics product to feature stacked, second generation High Bandwidth memory technology “HBM2”.
Vega 10 is aimed squarely at the high-end, enthusiast 4K gaming, VR & professional markets.
There’s been talk that AMD could showcase Vega 10 as early as this month. Although, official word remains that Vega is a 1H 2017 product.
AMD Graphics Architectures
|WCCFTech||Year||Product||GPU||Process||Transistors In Billions||Performance||Memory||Bandwidth|
|Southern Islands||2012||HD 7970 Ghz||Tahiti||28nm||4.3||4.1 TFLOPS||3GB GDDR5||264GB/s|
|Volcanic Islands||2013||R9 290X||Hawaii||28nm||6.2||5.6 TFLOPS||4GB GDDR5||320GB/s|
|Caribbean Islands||2015||R9 Fury X||Fiji||28nm||8.9||8.6 TFLOPS||4GB HBM1||512GB/s|
|POLARIS||2016||RX 480||Polaris 10||14nm||5.7||5.2 TFLOPS||8GB GDDR5||256 GB/s|
|VEGA||2017||RX Vega 64||Vega 10||14nm||12.5||13 TFLOPS||8GB HBM2||480GB/s|
|VEGA||2018||TBA||Vega 20||7nm||-||-||32GB HBM2||1 TB/s|
|NAVI||2019||TBA||Navi 10||7nm||-||-||Nextgen Memory||-|
With that being said AMD hasn’t been particularly shy about teasing some Vega tidbits this year. The new graphics architecture was first announced earlier this year during the company’s Capsaicin press event. Where Raja Koduri the ,then newly appointed, head of the Radeon Technologies Group laid out AMD’s GPU roadmap going all the way to 2018.
Since then we have actually spotted the fabled RX 490 card on AMD’s & Sapphire’s official websites. The card was touted as a 4K gaming capable high-end offering with a 2016 release date. Not long after that Raja Koduri announced that the Radeon Technologies Group was celebrating a major milestone with regards to Vega 10’s development. Subsequently, the head of PR at the Radeon Technologies Group teased the launch venue for Vega 10 on his facebook page.
Last we heard we could get a glimpse of Vega 10 for the very first time at AMD’s “New Horizon” event on December 13th.
The Key To A 50/50 Discrete Graphics Market Share
AMD CEO Dr. Su highlighted the company’s relationships with its customers as well as Vega’s competitive hardware to be the key market share expansion drivers going forward.
John Pitzer, Semiconductor Analyst at Credit Suisse :
How should we think about your share aspirations with Polaris now ramping, Vega next year? What do you think you can get your share back to within that market?
Dr. Lisa Su, AMD President & CEO :
We have made very good progress I would say in the first couple of quarters this year in terms of graphics. I think it’s nice, when you look at graphics it’s both in the consumer side on the channel as well as in the OEM business and on the professional graphics market. There’s a large opportunity. I think we’ve gained a good amount of share over the last few quarters, we’re going to continue and consistently drive [share growth]. We believe that there’s no reason we can’t be at 50/50 share overtime, but it will certainly take some time to get there. The key thing is enhancing our relationships with customers, because we believe that it’s also important to have a very sticky business going forward.
[…]I think the market is on-fire. It’s a great market. Our strategy in GPUs is to play across the entire market. We think we have the IP, the capability, the talent to do that. We started with gaining share in consumer and that’s why Polaris is so important for us this year. As we go into more of the higher-end markets with our next generation Vega architecture what you will find is the hardware is very very competitive.
This transcript is property of WCCFtech.com, any usage without appropriate citation is strictly prohibited.
The Graphics Market Is A Perpetual, Grueling Tug Of War
The last time AMD came close to a 50/50 market share position against its principal discrete graphics rival ,Nvidia, was back in 2010. The company had beaten its competitor to market with the first 40nm graphics chip the earlier year and followed with the highly successful HD 5870. Which was the most powerful and most power efficient graphics card at the time. Eventually securing it 51.1% of the discrete graphics market in Q2 2010.
|WCCFtech / Q2 2010||Desktop GPU Share||Mobile GPU Share||Overall Share|
Nvidia caught up a few months later with its GTX 400 series GPUs. Despite being significantly more expensive to produce and consistently more power hungry than the HD 5000 series, the GTX 400 series did just enough to tip the market share scales back in Nvidia’s favor.
The playing field today is quite different. AMD does not have the same technology advantage it did back in 2010. As things stand today, Polaris GPUs aren’t as power efficient as their Pascal counterparts nor are they as area efficient, making them more expensive to produce. Looking back at 2010, today AMD finds itself sitting at the opposite end of the table.
AMD’s GPU Share Challenge Is Not Without A Silver Lining
The company is now fighting an uphill battle. It needs Vega to be successful in closing the power and area efficiency gaps with Pascal. The latter is more important to ensure profitability. Smaller chips that perform just as well as their larger counterparts are cheaper to produce and can be sold for just as much making them more profitable. Which is why Nvidia has had great success expanding its profit margins. It simply makes more money on each Pascal chip it sells than AMD does on Polaris.
Perhaps the silver lining in all of this is that AMD’s current position permits it to drive the share momentum back in its favor by simply competing at the high-end . There are currently no Radeon counterparts to Nvidia’s GTX 1070, 1080 & Titan X offerings. By just giving customers a Radeon option at the high-end with Vega could do a lot to help the company’s position.
Even if Vega isn’t successful in completely closing the gap with Pascal. Simply offering a better performance per dollar deal is very appealing to customers. In fact it has proven time and time again to be the single most important metric by which a graphics card’s success is measured and ultimately defined. And it’s this key dynamic that AMD is betting on with all of its chips.