Monday, August 24th 2015
AMD Showcases Graphics, Energy Efficient Computing and Die-Stacking Innovation
Top technologists from AMD are detailing the engineering accomplishments behind the performance and energy efficiency of the new high-performance Accelerated Processing Unit (APU), codenamed "Carrizo," and the new AMD Radeon R9 Fury family of GPUs, codenamed "Fiji," at the prestigious annual Hot Chips symposium starting today. The presentations will focus on new details of the high-definition video and graphics processing engines on the 6th Generation AMD A-Series APU ("Carrizo"), and the eight year journey leading to die-stacking technology and all-new memory architecture included on the latest top-of-the-line AMD Radeon Fury Series GPUs ("Fiji") for 4K gaming and VR. Using a true System-on-Chip (SoC) design, 6th Generation AMD A-Series processors are designed to reduce the power consumed by the x86 cores alone by 40 percent, while providing substantial gains in CPU, graphics, and multimedia performance versus the prior generation APU. The new AMD Radeon R9 Fury X GPU achieves up to 1.5x the performance-per-watt of the previous high-end GPU from AMD.
"With our new generation of APU and GPU technology, our engineering teams left no stone unturned for performance and energy efficiency," said Mark Papermaster, chief technology officer at AMD. "Using innovative design for our APUs, we've vastly increased the number of transistors on-chip to increase functionality and performance, implemented advanced power management, and completed the hardware implementation of Heterogeneous System Architecture. For our latest GPUs, AMD is the first to introduce breakthrough technology in the form of die-stacking and High-Bandwidth Memory. The results are great products with very large generational performance-per-watt gains."
The details will be shared in two symposium presentations. On August 24, Guhan Krishnan, AMD fellow, design engineering, will present "Energy efficiency in graphics and multimedia in 28nm 'Carrizo' APU." This session will provide an in-depth view of the multitude of advancements resulting in superior performance, battery life, and user experiences on performance notebook and convertible form factors. In "AMD's next Generation GPU and memory architecture" on August 25, Joe Macri, corporate vice president and product chief technology officer at AMD, will share the journey from inception to market that covered eight years and involved several key partners, as well as provide architectural details behind the performance and efficiency of the new AMD Radeon R9 Fury line of GPUs.
Die-Stacking Journey Culminates In High-End GPU
The path to the new "Fiji" family of AMD Radeon R9 Fury GPUs began with exploring the best die-stacking option for bringing large amounts of memory into the same chip package with the GPU while dramatically increasing the memory bandwidth available to a high-performance graphics engine, without increasing power consumption. Working with memory partner SK Hynix, the new GPUs based on AMD Graphics Core Next (GCN) architecture offer up to 4 GB of high-bandwidth memory (HBM) over a 4096-bit interface to achieve an unprecedented 512 Gb/s memory bandwidth. The new memory is stacked close to the GPU in the package by implementing the first high-volume interposer as well as the first through-silicon vias (TSVs) and micro-bumps in the graphics industry. HBM and the interposer provide 60 percent more bandwidth than previous generation GDDR5 memory3 and 4x the performance-per-watt of GDDR5.4 At the same time, the "Fiji" family is capable of up to 8.6 TFLOPS performance, a nearly 35 percent increase over the previous generation (Radeon R9 290 Series GPUs).The result is an improvement of up to 1.5x performance-per-watt.
"With our new generation of APU and GPU technology, our engineering teams left no stone unturned for performance and energy efficiency," said Mark Papermaster, chief technology officer at AMD. "Using innovative design for our APUs, we've vastly increased the number of transistors on-chip to increase functionality and performance, implemented advanced power management, and completed the hardware implementation of Heterogeneous System Architecture. For our latest GPUs, AMD is the first to introduce breakthrough technology in the form of die-stacking and High-Bandwidth Memory. The results are great products with very large generational performance-per-watt gains."
The details will be shared in two symposium presentations. On August 24, Guhan Krishnan, AMD fellow, design engineering, will present "Energy efficiency in graphics and multimedia in 28nm 'Carrizo' APU." This session will provide an in-depth view of the multitude of advancements resulting in superior performance, battery life, and user experiences on performance notebook and convertible form factors. In "AMD's next Generation GPU and memory architecture" on August 25, Joe Macri, corporate vice president and product chief technology officer at AMD, will share the journey from inception to market that covered eight years and involved several key partners, as well as provide architectural details behind the performance and efficiency of the new AMD Radeon R9 Fury line of GPUs.
Die-Stacking Journey Culminates In High-End GPU
The path to the new "Fiji" family of AMD Radeon R9 Fury GPUs began with exploring the best die-stacking option for bringing large amounts of memory into the same chip package with the GPU while dramatically increasing the memory bandwidth available to a high-performance graphics engine, without increasing power consumption. Working with memory partner SK Hynix, the new GPUs based on AMD Graphics Core Next (GCN) architecture offer up to 4 GB of high-bandwidth memory (HBM) over a 4096-bit interface to achieve an unprecedented 512 Gb/s memory bandwidth. The new memory is stacked close to the GPU in the package by implementing the first high-volume interposer as well as the first through-silicon vias (TSVs) and micro-bumps in the graphics industry. HBM and the interposer provide 60 percent more bandwidth than previous generation GDDR5 memory3 and 4x the performance-per-watt of GDDR5.4 At the same time, the "Fiji" family is capable of up to 8.6 TFLOPS performance, a nearly 35 percent increase over the previous generation (Radeon R9 290 Series GPUs).The result is an improvement of up to 1.5x performance-per-watt.
13 Comments on AMD Showcases Graphics, Energy Efficient Computing and Die-Stacking Innovation
So could you please AMD, start doing something about that? PLEASE? Create a reference laptop design and give it for free to the Chinese companies. Dell and the others are still in bed with Intel.
I wish success because it's time for a good competition.
I can't wait to see APUs with HBM, using slow dual channel DDR3 off the system is a huge bottleneck of non-discrete laptop GPUs. It doesn't help when the iGPU will only use 512MB of RAM either (in my case) I just really hope it's not going to be incorporated in to the most expensive laptops, if so IMO it would make this less compelling. Give me budge laptops with HBM pls.
I'm assuming that it's the gaming performance you're after. Then we might look towards the chunkier, gaming-oriented laptops, as opposed to the ultrabooks, which would necessitate Carrizo-L due to its 15W TDP, and would not be a good fit for Carrizo-L because it would be a significant step-down in performance compared to the likes of the i5-5200U. These laptops are the ones that pack 30W+ CPUs and discrete graphics. No 720p screens? Alright, we'll axe everything below $900 from the list, the usual AMD territory for 720P, APU-powered laptops. What are we left with? A 1080p 15.6" or 17" laptop that probably is thick enough to dissipate dGPU heat and pack an optical drive, coming in at a price of around $1100. For around $1000-1100, all I see are MSI Gaming laptops packing a ULV / Mobile Broadwell i5 with Geforce GTX 950M graphics. In all likelihood, performance won't be too far off that of a Kaveri FX-7600P, but should beat the Kaveri by a small margin.
Now, Carrizo's main focus was on power efficiency, so we won't be factoring in any boost in CPU or GPU performance over Kaveri's top dog. A 1080P screen pretty much guarantees a price tag of around $1000 or above, since the 720P laptops occupy the space below that mark. You see, it isn't so easy to sell such a product. At that price or for a little bit more, we enter Nvidia's midrange-high end mobile graphics segment, which outperforms AMD's offerings, and the few Radeon mobile graphics SKUs in that region also outperform the APU. "But it's just an APU laptop with a 1080p screen, can't it go for $700-800?" It can't because the manufacturers have already established what an $800 laptop looks like, and even if it does have an AMD APU, it won't have a 1080p screen.
There are laptop models that attempt to introduce AMD into the above-budget market, but they all do stupid things like pair an APU with a Radeon graphics SKU that is barely above the APU's own iGPU in performance. It's no longer about Dual Graphics either.
It's not that I don't want to see an APU in something like a lower-spec Zenbook; I do. It's just not happening (because Carrizo-L doesn't make sense and) because the manufacturers are having a hard time justifying their use of AMD's past two APU families, Kaveri and Carrizo. Before that, Trinity and Richland were power hungry, but offered decent performance at 720p. So, in they went into $500-800 budget laptops that could do a surprising amount of gaming. Now, AMD wants to trade a bit of performance for power efficiency to expand into lower wattage market segments. The problem is, performance hasn't increased at all, and things like Carrizo-L leave us all scratching our heads. Carrizo-L isn't for ultraportables, but neither does it offer good-enough performance on the CPU side. So where does it belong?
Spelling is not my strong side.