What's new
  • Please do not post any links until you have 3 posts as they will automatically be rejected to prevent SPAM. Many words are also blocked due to being used in SPAM Messages. Thanks!

AMD Trinity: Going Mobile with a New APU

Status
Not open for further replies.

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
There should be no doubt in anyone’s mind about the current processor market’s direction. Instead of using multiple chips spread across a common platform, more and more core functionalities are being condensed into a single, all inclusive unit. Even in the smartphone market, every company is looking for the inherent efficiency and adaptable performance aspects than a System on a Chip or SoC architecture can bring to the table. After first integrating the memory controller and other Northbridge functions onto their CPU dies, AMD’s effort in this area eventually received a fitting name: Fusion

After years of championing the idea of incorporating a CPU and graphics subsystem onto a single chip, AMD’s Fusion goals were finally realized last year. The Brazos, Lynx and Sabine platforms serving the ultra portable, desktop and mobile platforms respectively, burst onto the market and found the public waiting with open arms. According to AMD, the age of the all inclusive compute unit had arrived and judging from their competitors' reactions, it’s hard to fault their vision. For example, Intel move their graphics controller on-die and as the recent Ivy Bridge launch showed, they’re quickly accelerating their performance objectives in this field.

Trinity has been created to maintain AMD’s commanding lead in the heterogeneous computing field but as with Llano, they’ve added a bit of a twist. Instead of focusing solely upon beating Intel from an x86 standpoint, this new architecture is meant to strike a delicate balance between the CPU cores and graphics pipeline. But make no mistake about it; the standard processing capabilities of the new Trinity APUs has been significantly improved over the previous generation. The GPU portion has also received a noticeable bump in performance in order to keep it heads and shoulders above anything Intel can currently offer.


Trinity may seem to be a close sibling to the outgoing Llano architecture but there’s more going on here than what first meets the eye. It uses a slightly larger die, houses more transistors, and is fabricated on the same manufacturing node but through advances in the 32nm HKMG process, performance per watt has been increased by an order of magnitude. As such, a new 17W, low voltage, high efficiency tier has been added to the APU lineup while the 35W, 65W and 100W units have remained standing but now feature additional performance benefits. AMD hopes this means traditional desktop and notebook form factors will speed up their transition towards sleek all in one systems and ultra slim, performance oriented mobile systems sporting extended battery life.

Long accused of releasing inefficient architectures, AMD is striving for performance per watt leadership, even though they won’t be making use of an advanced 22nm tri-gate manufacturing process like Intel. In order to do this, a focus has been put upon offering up the best possible blend gaming and entertainment experiences, two areas where traditional processors fall woefully short.


Initially, the Trinity lineup will be solely geared towards the mobile market, an area we’re sure it will excel in. AMD’s new A10-4600M heads things up with a quartet of cores with a maximum frequency of 3.2GHz through the use of Turbo Core 3.0, 2MB of L2 cache and a TDP of just 35W. When compared against the previous generation’s top end Llano A8, the improvements are certainly there on every front.

Alongside the A10, AMD will be releasing the A8-4800M and A6-4400M, two processors which also have a TDP of 35W but should consume significantly less power. The 4800M still uses four cores and 4MB of L2 cache but its clock speeds have been reduced and the integrated GPU's SIMD array has been cut down to 256 cores. The 4400M is a dual core processor with a mere 512KB of L2 cache per core and a lower end graphics stage but its clock speeds match those of the higher end A10.

Further down-market are the low voltage and ultra low voltage APUs for Ultrabook-like mobile platforms. We already saw their introduction within HP’s new Sleekbook, a great looking Ultrabook competitor that promises huge amounts of battery life. The A10-4655M and A6-4455M are similar in layout to their full voltage cousins but due to decreased operating voltage, clock speeds have been severely curtailed in favor of lower TDP values.

One of the major changes within Trinity's architecture is the use of Radeon graphics cores featuring the Northern Islands architecture (otherwise known as the HD 6000-series) and full support for DX11. The number of graphics processing engines has been actually been decreased over the HD 5000-series Evergreen-based cores within Llano but that doesn’t mean less performance as the VLIW4 architecture holds a significant core per core lead over VLIW5. An expanded feature and functionality set has been built in as well.

While we’ll be getting into the gritty details of this architecture and it’s impressive potentially, you’ll have to wait a few weeks for a full Trinity review from us. In addition, today marks the introduction the Trinity mobile platform, code named Comal while the desktop-oriented Virgo platform will see its launch in the coming months.

 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Trinity’s Position in Today’s Market

Trinity’s Position in Today’s Market


Currently, the notebook and desktop market are dominated by Intel’s Sandy Bridge and Ivy Bridge architectures. AMD mounted a valiant effort last year with their Llano and Bulldozer architectures but ultimately fell short of their goals and some people’s expectations. The new 2012 lineup on the other hand is supposed to bring their products into a more competitive position against alternatives from Intel.


AMD’s upcoming product stack should look familiar to you since it virtually mirrors last year’s, with the new A10-series APUs being the sole exception. As usual, the eight core FX-series –with updated Piledriver cores- will take the lead in high end desktop and mobile gaming systems. The A10-series of APUs will follow closely behind with four cores, support for the open source HD3D stereoscopic 3D environment and will represent the highest end designation for Trinity-series APUs. Both of these products are meant to deliver outstanding performance without costing a fortune.

Going slightly down-market we have the A8 and A6 APUs which continue the tradition of Llano by offering affordable options in the mainstream notebook and desktop markets. Naturally, they’ll offer a full array of features but in a more efficient package. With the exception of the A4-series, all APUs will have support for Turbo Core, Dual Graphics and in some cases (for desktop systems only) will have the “K” branding, allowing for unlocked overclocking.

The A4-series doesn’t have as lofty goals but it should still be considered a feature rich architecture with support for up to four displays, AMD’s HD Media Accelerator and a number of GPU compute add-ons like Steady Video 2.0 technology. For the time being, AMD isn’t officially announcing any A4 APUs but expect them to make it to market within the next few months.


Unfortunately, AMD was not able to provide us with a non-watermarked image

Finally, in a trailing position is the new Brazos 2.0, an encore presentation of the original Brazos. It improves upon its predecessor in a number of key areas but retains the same Zacate-based architecture with some key revisions. There’s a faster HD 7000-series graphics subsystem (which is actually a rebranded HD 6000-series core), slightly higher clock speeds and a number of connectivity additions which bring Brazos 2.0 up to a standard that reflects current market realities. Most importantly, power requirements have been optimized, resulting in battery life that can exceed 10 hours in some cases. For anyone looking for a sub-notebook, this is one you’ll want to watch out for.


We mentioned in our long-winded introduction that AMD’s isn’t trying to compete against higher end Intel offerings. As such, the flagship A10 series’ x86 performance will likely land somewhere between the i7 and i5 processors while the A8 APUs will target the i5/i3 market position. The A6 and A4 meanwhile are supposed to take advantage of a perceived gap between the Core i3 and Pentium products.

Naturally, the Brazos 2.0 E1 and E2-series round out the new lineup and will concentrate upon the entry level price points. The only constant between AMD’s outgoing and incoming product stacks is the C series which marks a continuation of the older Brazos 1.0 Ontario-based architecture and should compete against Intel’s nearly defunct Atom series.

AMD’s Trinity lineup may not represent a new, game changing level of x86 performance, nor is it particularly efficient when compared against similar Sandy Bridge and Ivy Bridge products but these new APUs do offer a distinct price advantage for OEMs. For example, the HP Sleekbook we talked about will sell for about between $599 and $699 while most Intel-based Ultrabooks easily hit the $999 mark. When combined with Trinity’s expanded feature set and enhanced graphics capabilities, AMD could be in the process of moving the entire mobile market forward by a few steps.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Revising Bulldozer; Say Hello to Piledriver

Revising Bulldozer; Say Hello to Piledriver


Some of you may be wonder what a Piledriver (the artist formerly known as Enhanced Bulldozer) section is doing in an APU review but there’s a reason behind our…errr…AMD’s madness. Instead of sticking to the old Athlon / Husky CPU architecture that graced the Llano APUs, AMD has now moved on towards higher performance CPU cores that are based off of a revision of the modular Bulldozer design. Code named Piledriver, we’ll be seeing this optimized design in dedicated eight core CPUs sometime in the future but for the time being, it has been included in Trinity.

Naturally, the inclusion of Piledriver CPU cores should make a night and day difference in terms of x86 performance over the previous generation but there are some tertiary benefits as well. Turbo Core 3.0 has been incorporated, resulting in better efficiency through highly adaptable clock speed scaling and instruction paths have been optimized.


Based around a 32nm manufacturing process, equipped with a pair of cores and up to 2MB of L2 cache, the basic Bulldozer module hasn’t changed all that much in its Piledriver guise. There have been some minor changes like the addition of the FMA3 and F16C instruction sets but AMD’s major focus here was to increase the instruction per clock (IPC) rate and generally improve upon the operational frequencies of the previous generation. The relative maturity of GlobalFoundries’ 32nm node also led to a substantial leakage reduction when compared against Bulldozer.

For those of you keeping track of such things, there’s a huge difference between Llano’s Propus / Husky core architecture and this one. We won’t bore you with fine grain details here (more about the Bulldozer architecture can be found HERE) but the move to Piledriver compute modules has resulted in a claimed 29% performance increase for the mobile platform.


Diving a bit further into the Piledriver, the enhancements seem to be everywhere. Most are supposed to home in on streamlining branch scheduling throughout the architecture and optimize certain elements for quicker communication. We’ll cover more of this when Piledriver launches on the desktop so for the time being, jump to the next page to see how this change in core module design affects the Trinity APUs.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Under the Hood: Trinity’s Architecture

Under the Hood: Trinity’s Architecture



As we have mentioned previously, the basic layout of Trinity isn’t that far removed from the first generation of APUs but there are some significant changes. In its highest end A10 configuration, Trinity will include up to two Piledriver compute modules with two cores each and 2MB of L2 cache for quick data access. Lower end derivatives will only include a single module with two cores and in some cases only 512KB of memory per core, thus reducing power consumption even more.

The two 64-bit memory controllers have also been updated to support a broader range of P-states which is particularly important for mobile platforms since it allows for on the fly memory frequency scaling. There’s also built in support for 1.25V DIMMs but like in Llano, the controllers still support up to 32GB for notebooks and 64GB in desktop systems in dual channel mode.


The major changes in the Trinity architecture are buried within the new Unified Northbridge which represents the AMD’s first attempt at creating an all-in-one communication solution for their present and future APUs. Within it, a dedicated PCI-E link replaces Hypertransport protocol to the chip’s main I/O devices, APU power management can be regulated on the fly and memory controller requests can be effectively shared between the x86 processing stages and the GPU.


The links between each section of the APU follow in the same footsteps as the previous generation but AMD has refined certain interconnects with the goal of speeding up information transfers. The AMD Fusion Compute Link is still considered to be a medium bandwidth connection which manages the complex interaction between the onboard GPU, the CPU’s cache and the system memory. Unlike in the past, AMD has finally refined this interconnect, giving the GPU direct access to a coherent memory space while the CPU can now directly access the GPU’s dedicated framebuffer if needed. This is one of the primary reasons why Trinity’s theoretical data throughput has jumped from 572 GFLOPS to 736 GFLOPS.

The Radeon Memory Bus on the other hand is the all-important link between the onboard graphics coprocessor and the primary on-chip memory controller. Rather than acting like a traffic cop (a la Fusion Compute Link) which tries to direct the flow of information, this memory bus is all about the GPU having unhindered high bandwidth access to the system’s memory controllers.

In the previous generations of AMD IGPs, before Llano came around, the Northbridge’s graphics processor had to jump through a series of hoops before gaining access to onboard memory which is partially why 128MB of “SidePort” memory was sometimes added. However, the APU’s single chip, all in one solution allows for the elimination of many potential bottlenecks.


The graphics core within Trinity should look familiar since we last saw this payout back in the HD 6000 days. Instead of using the newer GCN Southern Islands architecture, Trinity’s SIMD engines rely on the slightly older Northern Islands with its VLIW4 instruction set. The only exception is the new Video Codec Engine which acts as one stop shop for hardware encoding via the GPU’s compute engine and provides a highly parallel scalable pipeline for many high definition tasks. It can also provide additional benefits for transcoding and output tasks.

Even though Trinity uses last year’s GPU architecture, it still comes equipped with some serious graphics processing muscle. In an A10 APU, there will be 384 stream processing units (or cores), 24 texture units and 8 ROPs, mirroring the layout of a HD 6600-series desktop part.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Reducing Power to Prolong Battery Life

Reducing Power to Prolong Battery Life


Thermal and power efficiency are integral parts of any processor architecture, particularly in the mobile area. Intel has addressed this through the progressive introduction of smaller manufacturing nodes but AMD doesn’t have the luxury of advanced 28nm or 22nm production just yet. This has necessitated some innovative thinking on their part. Naturally, the relative maturity of GlobalFoundries’ 32nm HKMG "gate first" technology has gone a long way towards optimizing Trinity’s power delta but there are some additional technologies working behind the scenes too.


Throughout the years, AMD has been gradually implementing measures to ensure their processors remain as efficient as possible. From fine grain power gating to an integration of tertiary components onto a single die, the progression has been slow but constant.

Llano introduced us to a number of dynamic thermal and power management techniques which were meant to ensure it didn’t exceed certain predetermined limits. Unfortunately, the implementation was a rather rudimentary one, which the Trinity architecture is now aiming to improve upon. While Trinity still includes power gating and thermal management technologies, it now features adaptive, situation aware voltage regulation. In plain English, this allows it to dynamically deliver performance when needed, only engaging the APU’s full capabilities when needed. Idle power has also been reduced to an incredible 1.08W.

This is just the tip of the iceberg since AMD is hard at work researching additional power saving techniques as well. One of these is the Heterogeneous System Architecture which will usher in a whole new round of architectural communication streamlining. This will go a long way towards improving overall APU efficiency due to its ability to prioritize processing tasks towards the coprocessor that’s best able to carry out the command. Even though AMD’s HSA efforts will start in earnest next year, Trinity represents a stepping stone towards enhanced integration and efficiency for APUs.


We mentioned in the Llano review that AMD had implemented a simplified power transfer protocol between the CPU and GPU once the SIMD array’s activity was low. This approach allowed for a basic balancing of TDP loads in order to ensure the APU operated within a predetermined area of thermal and power constraints.

Trinity on the other hand is able to dynamically balance the operating points of the GPU SIMD engines and each of the x86 CPU cores in order to optimize the architecture’s throughput. This allows for both processing stages to work in parallel without negatively affecting overall performance and should result in fine-grain control over efficiency, thus increasing battery life. In essence, this allows both the CPU and GPU to operate at “base” and “turbo” frequencies provided each stays under AMD's TDP value.


With power gating, updated silicon and other power management technologies, AMD’s mobile Trinity platform (called Comal) should now be able to deliver up to 10 hours of battery life. More importantly, the careful integration of peripheral components onto the APU’s die has allowed for a substantial savings in total system power needs as well.


According to AMD’s figures, their new AllDay power scheme will result in significantly longer battery life. As you can see above, the optimized Piledriver CPU cores, more efficient power delivery and a number of other advances make a top-tier A10-4600M last just as long or longer than Intel’s mid-level i5-2410M. In addition, the areas where the APU looses still represent somewhat of a “win” since it is able to deliver substantially more performance in HD decoding and video streaming.


In comparison to Llano, Trinity’s advances meant it is able to achieve longer battery life through nearly every scenario, even though its performance is much improved. AMD is hoping this will make their new architecture appealing to system designers and improve upon Llano’s design win success.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Taking Gaming to the Next Level

Taking Gaming to the Next Level


When Llano was first introduced, we were rightly impressed with how much graphics processing power AMD was able to cram into such a limited die area. Unfortunately, the graphics processor within Trinity may use the HD 7000-series designation but its architecture stems from the Northern Islands or HD 6000-series family. We’ll likely see a closer integration of the current Graphics Core Next architecture into upcoming APUs, where its advanced feature set can be faithfully brought into the HSA fold.

While the HD 6000-series may be yesterday’s news in the desktop market, it is still based off of a highly adaptable core design that offers more than enough power for the notebook market. AMD has harnessed these abilities by cramming an impressive number of processing cores into every corner of their Trinity APUs, making them a quantum leap forward in terms of overall graphics performance. But just how much? We’ll let some of AMD’s internal slides do some talking for us.




While the slides above surely paint a rosy picture for the future of Trinity, there are some small speed bumps that may be placed in Trinity’s way. Since the performance results were published about a month ago, AMD deftly avoided the three ton guerilla in the room: Ivy Bridge. The real question here is whether Ivy Bridge’s distinct advantage in x86 processing will allow its in-game performance to partially catch up with AMD’s upcoming Trinity APUs. We highly doubt it, but the spread will be much closer than what’s evident above.


We’ll end this section off with a desktop-oriented slide since it is the only one available that clearly shows Trinity’s performance in relation to Llano. Naturally, the 5800K could have been matched up with the A8-3870K (its clock speed is 100MHz higher than the 3850) but the trend is clear as day: from a gaming perspective, Trinity’s advantage over the outgoing generation is clear as day, and this will translate directly into the mobile market as well.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Dual Graphics Rises Again

Dual Graphics Rises Again


AMD’s Dual Graphics technology hasn’t been talked about all that much but it really does deserve a spot in this article. Introduced with Trinity’s predecessor, Dual Graphics allows an APU to be paired up with a discrete graphics chip for increased performance. Think of this as a type of hybrid Crossfire where similar GPUs of the same family can communicate with one another when necessary.


Even though the notebook manufacturers won’t necessarily allow the end user to pick and chose their poison, every one of the Trinity APUs can be combined with a discrete solution, resulting in a nearly endless number of options. The chart above represents only a small cross section of what’s possible.

For those of you worried about power consumption, the discrete card will go into a state of suspended sleep until it is needed, thus having as little impact as possible upon battery life.


Dual Graphics incorporates many of the same features as Crossfire, meaning performance increases can vary from one application to another but on average, scaling is should be quite good. Supposedly, an A10 APU combined with a lower end discrete chip can easily equal the numbers achieved by an upper mid tier discrete card, without the associated thermal impact. Application profiles have also been instituted so optimizations can be rolled out to Dual Graphics enabled systems without having to worry about the vendors’ driver stack.


Battlefield 3 is one of the most popular games around and it seems we can’t go through a single review without at least mentioning it. AMD claims that with a single A10 APU, you should be able to play it on medium detail settings but when combined with a HD 7650M discrete add in card, performance can be substantially increased without the notebook in question becoming unaffordable.
 
Last edited:

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Broadening the Horizons for GPU Compute

Broadening the Horizons for GPU Compute


For the last few years, we’ve all heard the “compute!” cry from the AMD and NVIDIA camps and more recently, a growing list of software developers have taken up the cause. The entire focus behind AMD’s Accelerated Processing Units hinges upon a closer integration of GPU computing alongside CPU processing so it is naturally in their best interest if the compute ecosystem expands. This is particularly important for upcoming APU architectures which will leverage the new HSA approach in a big way.


It hasn’t taken long for GPGPU-supporting software to get up and running. Many developers have quickly embraced the idea of their applications being granted additional parallel processing cabilities of modern graphics architectures. As a matter of fact, AMD’s VLIW4-based SIMD arrays within the Trinity architecture (and to a greater extent the GCN architecture within upcomin APUs) has been partially designed around the possibility of delivering optimal compute performance without taking a chunk out of battery life.

To accomplish this, AMD is leveraging OpenCL in a big way and considers it a win / win solution for developers and consumers. The API gives developers access to an open GPGPU standard versus an expensive licensed solution, thus allowing them to leverage higher performance though the use of parallel processing. On the other hand, consumers will benefit through improved application / hardware detection and faster task completion times.


At the heart of any compute initiative rests the ability to encode or transcode videos for use on today’s devices. AMD’s GPU Accelerated Video Encoder features support for multi stream H264 HD encoding and leverages the APU’s built-in processing stages for enhanced performance. There are also several conversion options available and we’ll go through all of them within our complete review.


Speaking of expanded support for GPU parallel processing and the benefit of APUs in these situations, WinZip has recently introduced support for OpenCL. As a result, AMD’s Trinity architecture is able to shine, supposedly delivering a noticable performance improvement over the previous CPU-centric version. Considering this support is also offered on the free version of WinZip, a large cross section of the users could now benefit from faster compression times.


MotionDSP’s vReveal software has been tied at the hip to GPU compute since its inception and naturally, the Trinity architecture allows for significant performance benefits. Again, this is due to its accelerated video encoder stepping to the forefront.


GIMP, everyone’s favorite (and free!) image editing software now has OpenCL support for GPU accelerated tasks such as rendering and color conversion. Using a fraction of the onboard GPU’s power for these tasks results in higher performance without a decrease in system efficiency.


GIMP may be a freeware program but even the ultra expensive haven’t been spared from the ongoing wave of GPU accelerated acceptance. As many are aware, the latest version of Photoshop fully supports GPU compute algorithms, potentially making any Trinity-based notebook an optimal solution for on the go professionals.

This section just scratched the surface of what GPU compute can bring to the table and over the course of 2012, we’ll surely see more applications take advantage of it. Luckily, AMD’s forward thinking mentality in this field has all but assured Trinity’s preeminence….for the time being at least.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
An Entertainment Powerhouse

An Entertainment Powerhouse


As the market slowly shifts away from desktop-centric system, more and more importance is being put upon the capability (or lack thereof) of mobile platforms to provide a rich entertainment experience. Previous architectures struggled to deliver high quality video streams and simply couldn’t provide the decoding performance necessary for stereoscopic 3D signal processing. Trinity on the other hand takes the improvements from the Llano architecture and improves upon then in a few key areas.


In terms of HD decoding, Trinity’s use of UVD3 doesn’t bring anything new to the table when compared to Llano other than additional performance and enhanced efficiency. This is completely understandable since Llano already had an extensive list of supported codecs and features.

Remember, AMD’s Universal Video Decoder has been around for years and is known as one of the most capable video processing platforms currently available. In its second iteration, UVD took the next logical step forward with an expanded list of accelerated codecs in addition to the ones already in place from past generations.

One of the main features which was added to UVD3 was the ability to decode videos which use MVC encoding. As part of the H264 / MPEG-4 AVC codec, MVC is responsible for creating the dual video bitstreams which are essential for stereoscopic 3D output. Supporting this standard gives AMD’s APUs the ability to process Blu Ray 3D movies through a HDMI 1.4a connector. MPEG-4 Part 2 hardware acceleration for DivX and Xvid codecs has also been added.


While AMD has stayed the course on the HD decoding front, Trinity brings a long list of new output features due to its use of DisplayPort 1.2. DP 1.2 brings to the table multi-stream support for display daisy chaining and enhanced Eyefinity options for up to four simultaneous displays, both of which could come in handy for future AV applications. The new Vision Control Center allows for display grouping as well, giving users the possibility of combining multiple displays to act as a single large display. In addition, the GPU display engine has built-in wireless display compatibility.

Sound output options have been expanded with support for high bitrate 7.1 channel surround sound over HDMI and DisplayPort. This means audio formats such as PCM, AC-3, AAC, DTS, Dolby TrueHD and DTS MA are all included and to add icing on the cake, Trinity can process up to FOUR independent 7.1 audio streams.

As with many of AMD’s current GPU architectures, Trinity has several built-in image quality enhancements which include 4:2:0 color sampling, optimizations for scenes changes audio / video multiplexing and color gamut remapping for wide gamut panels.


All of the new entertainment features have been neatly packaged into a generalized Media Accelerator but this shouldn’t in any way reduce their effectiveness. There are quite a few elements here. AMD Picture Perfect HD boosts contrast, colour and resolution in real time for improved picture quality while Steady Video can be enabled in the Vision Control Panel and will automatically smooth out videos, be they online or from your own hard drive. We’ve already touched upon the Accelerated Video Converter but the most interesting new feature has to be AMD Quick Stream technology.


Quick Stream is an easy marketing name for a complex technology which allows for drastic performance improvements in certain predetermined types of online content. It prioritizes the downstream and upstream internet packets in real time in an effort to accelerate the most demanding workloads (like Youtube video streaming) without negatively impacting general usage scenarios like web surfing.

Unfortunately this option isn’t user controllable but it can be implemented by a manufacturer that wants to put emphasis upon certain online functions. For example, the manufacturer of a gaming notebook could give priority to Steam and online gaming traffic while minimizing the bandwidth used by Windows updates and general web surfing. Alternately, entertainment PCs may put higher emphasis upon Youtube video priority and other Flash / Silverlight / HTML5 streams.
 

SKYMTL

HardwareCanuck Review Editor
Staff member
Joined
Feb 26, 2007
Messages
12,857
Location
Montreal
Parting Thoughts

Parting Thoughts


Let’s start this off with some words of wisdom: while we haven’t been able to independently confirm AMD’s numbers, Trinity certainly has potential. It improves upon Llano in a number of key ways and could usher in a new age for compact, efficient, yet powerful mobile platforms.

While Trinity may be the next evolutionary step for the Fusion initiative, no one should mistake this as a game changing architecture. The use of Piledriver CPU compute modules, Northern Islands SIMD units and more efficient communication pathways will certainly bring some benefits to the table but we’re not expecting miracles here. Nonetheless, AMD is betting the farm on their belief that certain segments of the market don’t need high end x86 processing anymore, especially when the addition of a simple, relatively inexpensive SSD can make a system feel so much more responsive. And guess what? No matter how many people second guess their direction, AMD has a valid point: in the grand scheme of make-believe system “performance” in the mobile market, a faster x86 processor rarely makes a difference.

Striking a seamless balance between CPU and GPU isn’t an easy thing to do, particularly when much of Trinity’s potential hinges upon software support. About a year ago we mentioned that AMD would need developer support to successfully carry out their Fusion plans and that’s exactly what happened. While many other “open” initiatives seem to fail before being fully realized, OpenCL is finding adherents and finding them quickly.


Unfortunately, many of AMD’s comparisons we included in this article concentrated upon the Intel Sandy Bridge architecture and let’s be honest here; things will change dramatically when Ivy Bridge is fully available in the mobile product space. And make no mistake about it, widespread IVB availability on notebooks is only weeks away, potentially spoiling AMD’s party before it can even get started.

From a mobile perspective, Trinity is certainly headed in the right direction but they’ll need to pick up widespread OEM support in order to be successful. The HP Sleekbook is a prime example of this: it isn’t a “poor man’s Ultrabook” as some have loudly proclaimed. Rather, it looks like a fully featured, ultra portable, highly affordable system which should offer great performance without a meaningless price premium.

Trinity’s goal is simple: bring efficient, price conscious, balanced performance to consumers without the associated “Intel tax” some have been complaining about. While this may or may not translate well into the desktop market, we honestly hope that Trinity will prove to be a worthy adversary against Ivy Bridge in the mobile product space.
 
Status
Not open for further replies.

Latest posts

Twitter

Top