Sign in with
Sign up | Sign in
AMD FirePro W8100 Review: The Professional Radeon R9 290
By ,
1. Introducing AMD's FirePro W8100 Workstation Graphics Card

We already covered the Hawaii GPU's debut in the workstation space (AMD FirePro W9100 Review: Hawaii Puts On Its Suit And Tie). That story began with a question: did AMD throw caution to the wind and design a professional-class card designed for maximum rendering horsepower, rather than target a sweet spot? The company's approach changes with its FirePro W8100, so now we have to ask if the newer board is fast enough to warrant its asking price.

Where is AMD trying to go with the W8100? In the presentation slides for its W9100, the company set that card up as a competitor going against Nvidia's Quadro K5000, and was taken by surprise when it succeeded on all fronts. Based on the outcome, AMD became a bit more confident, and is now positioning the FirePro W8100 as the right card to go against the Quadro K5000. Its W9100 now shoots even higher.

What about pricing, you ask? The W8100 hasn't shown up for sale yet, but it's expected by the end of July at a price point of $2500 (compared to the K5000's $1800). Given those figures, AMD needs to hope its card still performs significantly better.

Spoiler alert! In the table above, we see that the FirePro W8100’s measured power consumption is approximately 28 percent lower than the W9100's. With a compute-oriented load applied, it draws noticeably less power than Nvidia’s Quadro K6000 for the very first time, and it is about on par with it in 3D tasks. This lower power consumption is roughly what you can expect, keeping in mind lower performance, when both figures are expressed as percentages. Of course, that doesn't mean our real-world benchmarks will yield the same findings, so the test results should be interesting.

Quo Vadis, AMD FirePro W8100?

When you have performance to offer, new opportunities present themselves. AMD identifies CAD and engineering, media and entertainment, medicine, and finance as some of the FirePro family's more traditional strengths. But with its big Hawaii GPU and the GCN architecture's alacrity in compute-intensive tasks, AMD wants to lock down its share of the virtualization, cloud gaming, and signage segments as well.

The ambition makes sense. Workstation-oriented apps benefit more and more from the performance of modern GPUs, after all. Nowadays, you can even run multiple CAD and CAE workflows at the same time. Cranking along on the next version of a drawing while rendering the previous one isn't a pipe dream. This stuff is actually doable. And the sky's the limit with a design equally adept in 3D- and general-purpose tasks.

AMD is already a seasoned vet when it comes to 3D. Now GPGPU is where it's trying to lead development. In order to better facilitate that initiative, the company is throwing its support behind the OpenCL standard as an alternative to Stream and CUDA. As we've seen in several different applications already, when there's a computationally difficult job that can be parallelized, the potential performance gains are well worth optimizing for.

There's also a notable trend toward the adoption of 4K (3840x2160) in the workplace. Those higher resolutions give engineers and artists a lot more room to work with. And while more detail obviously benefits 3D applications, even 2D tasks like programming are greatly enhanced by the extra screen space and pixel density of a 4K display.

Similarly, professional media-oriented titles see a lot of benefit as it becomes possible to edit high-res video in real time at full resolution. A workstation board like the W8100 should speed up the processing of video and photo filters, along with accelerating encoding/decoding. The professional graphics card market is clearly changing, and the lines between various segments are getting blurrier, even as the workloads and data sets are more specific than ever. CAD, CAE, M&E, oil and gas...the W8100 is AMD’s most recent effort to grab a larger share of all of them by further diversifying its portfolio of FirePro products.

AMD says that the FirePro W8100 is supposed to have a great price/performance ratio, and, in light of the card’s price, it could be onto something special. Is it the real deal though, faced with a less expensive Quadro K5000 as competition?

2. Dimensions, Weight, Features and Pictures

Pictures, Features, and Connectors

Let’s first take a look at the most important dimensions, and then move on to the pictures and features.

Dimensions and Weight
Length
282 mm, or >300 mm including the power connectors
(the PCIe power connectors are in the back)
Depth
34 mm from PCB to top of fans
5 mm from back of PCB to top of back plate
Height103 mm from top of PCIe slot
Weight1086 g

Just like AMD's FirePro W9100, the W8100 looks quite inconspicuous. Its plain black plastic cover reminds us of the Radeon HD 6970. Even the reference cooler appears to remain the same, which is somewhat disappointing next to Nvidia's redesigned Quadro cards.

The FirePro W8100’s thermal solution employs the same vapor chamber cooler we know from AMD's W9100 and W9000. The prominent red fan forces air through the cooler. Heated air is expelled through the left side of the card, out of its I/O bracket. As we already know from Radeon R9 290X Review: AMD's Back In Ultra-High-End Gaming, there is no way for this configuration to run quietly. But we are certain that it does its job.

The back of the card is dominated by a metal plate, which adds rigidity and does double duty cooling the memory packages mounted on that side of the PCB. There’s something new as well, though. Rubber spacers are glued onto the card to guarantee enough space between graphics cards in a multi-GPU setup.

There's not much to see on the bottom except for this card's closed shroud. Those rubber spacers are pretty obvious.

The top of the card doesn’t sport any CrossFire connectors. Remember, though, that the Hawaii GPU employs a DMA engine, which enables CrossFire support through the PCI Express bus.

There is one header along the top edge though, which is also present on the FirePro W9000. It's used for connecting AMD's FirePro S400 synchronization module.

Two six-pin auxiliary power connectors are found on the back of AMD's FirePro W8100, which is a change from the W9100's six- and eight-pin inputs. This reconfirms the new card’s lower power consumption.

Slot Panel Connectors

Four DisplayPort connectors drive up to a quartet of 4K panels at 30 Hz. Or, you can connect three 4K panels at 60 Hz. The Hawaii GPU comes equipped with six display engines though, so it's possible to attach an MST hub and attach that many screens at lower resolutions. There is also a three-pin mini-DIN connector for 3D displays.

3. How We Test AMD's FirePro W8100

Test Systems and Environment

For this story, we don't overclock Intel's Core i7-4930K, since the workstation world is very stability-sensitive. As a result, our processor runs at a base clock rate of 3.5 GHz. But this machine's test system does sport three SSDs now. We keep the operating system separate from the benchmark suite binaries and data logs.

Normally, we would only test with drivers approved by each ISV. However, this isn’t possible for a brand-new card, so we had to use the latest driver available for AMD's FirePro W8100 (refer to table).

The power draw measurements deserve a section of their own, and we're eager to find out if the AMD FirePro W8100 is really in the sweet spot of AMD's line-up.

Here's the list of hardware we're using for benchmarking:

Lab Bench
Microcool Banchetto 101
System
Intel Core i7-4930K (Ivy Bridge-E), 3.5 GHz (6C/12T)
Asus Rampage IV Black Edition
64 GB Corsair Dominator Platinum DDR3-2133 at 1600 MT/s
Corsair H105i Closed-Loop Liquid Cooler
1x 256 GB Samsung 840 Pro SSD (System, Applications)
Video Editing / Workloads
1x 480 GB Corsair Neutron GX SSD (Input)
1x 500 GB Samsung 840 EVO SSD (Output)
Power Supply (PSU)
1200 W be quiet! Dark Power Pro 10 (Slightly Modified to Measure Voltages)
Operating Systems
Microsoft Windows 7 Ultimate x64 (Professional Applications, Computing)
Microsoft Windows 8 Professional x64 (Gaming)
Drivers
Catalyst Pro 13.352.1009 Certified
Nvidia Quadro Desktop Driver 334.95
Environment
22 °C Ambient Temperature, Held Stable via Air Conditioning


4. OpenCL: Compute, Cryptography, and Bandwidth

Shader Performance: FP32 vs. FP64

Let’s start with an OpenCL benchmark, which should push the theoretical ceiling of 32- and 64-bit compute performance.

Sandra's Cryptography module is next. It’s remarkable how well the older FirePro W9000 keeps up. The distance between the W9100 and the W8100 is right around where it should be based on each card's specifications.

Although these benchmarks are synthetic in nature, they still illustrate Nvidia’s half-hearted support of OpenCL. Yes, the company offers its proprietary CUDA API, and there are plenty of applications that support it. Increasingly, though, ISVs looking for a broader customer base don't want to support two languages, and OpenCL is gaining traction as a result. Even long-time bastions of CUDA support like Adobe are adopting OpenCL.

Folding It Up: Folding@Home

Let’s run the Folding@Home benchmark on this card. Even though few professionals would use a workstation-class board for this (or cryptocurrency mining), the test does give us a more real-world look at compute performance.

Once again, the gap between each card's performance is what we'd expect in light of their specifications. This chart demonstrates nicely how well AMD's architecture scales in scenarios without overhead.

Memory Bandwidth

Conversely, when comparing memory bandwidth under OpenCL to Direct3D 11, Nvidia demonstrates that putting more effort into optimizing drivers makes a quantifiable difference.

As we move on to our application benchmarks, keep these synthetics tests in mind. They help decipher the performance results of real-world metrics, which are subject to influence from other platform subsystems.

At least for now, we have to question whether Nvidia's lackluster support for OpenCL and emphasis on CUDA is the best strategy. Only time will tell.

5. OpenCL: Financial Mathematics and Scientific Computations

OpenCL: Financial Mathematics

Option pricing is a compute-intensive task that gives graphics processors an opportunity to shine. Both synthetic OpenCL-based benchmarks are clearly dominated by AMD.

It does stand out, however, that the ratio between computations with single and double precision is much worse in some places than it is for Nvidia’s Quadro K6000. The K5000, on the other hand, isn't even a contender.

Scientific Computations

Looking at the GEMM benchmark’s single-precision performance results, the  numbers turn out as we'd expect.

The same can’t be said for computations employing double precision. All three AMD graphics cards suffer massive performance drops, which is somewhat hard to explain since AMD claims its FP64 rate is one half of FP32.

That halving of performance is more believable in our FFT test, which shows the FirePro W9100 and W8100 shedding enough FP64 throughput to allow Nvidia's Quadro K6000 to finish in first place.

6. 2D Performance: GDI and GDI+

Why Are We Still Looking At GDI and GDI+?

Even in 2014, many applications use GDI and GDI+ for drawing, even if only for their GUIs. Older productivity applications and specific business titles still leverage GDI/GDI+ predominantly. These applications range from simple 2D CAD programs and viewers to pre-print stage WYSIWYG layout programs and file import/export programs.

As modern graphics cards with unified shaders don’t feature dedicated 2D units anymore, and modern operating systems no longer access graphics cards directly, device drivers play a crucial role in facilitating fast 2D functions.

Text

Displaying text is a crucial task and, needless to say, both manufacturers make sure that their high-end graphics cards render large amounts of text almost instantly.

Lines

Another basic 2D element is lines (the lines in a menu, for example). Again, none of the cards encounter problems with this task, though we notice that AMD's products are around 20 percent faster than Nvidia's.

Splines / Bezier Curves

Curvy lines require some computational power, and it's only natural that they take longer to draw. Just as we saw in the line test, there's a noticeable gap between the AMD and Nvidia cards. The FirePro W9100 and W8100 come close to each other, even though the older W9000 does manage to slide in between them.

Polygons

This benchmark draws filled and unfilled polygons with three to eight vertices; AMD's hardware knows how to handle it. You can't say the same for Nvidia's Quadro K5000, which is slightly better than half as fast as the older FirePro W9000.

Rectangles

Yes, rectangles are polygons, too. But GDI exports a separate, simpler API for drawing them. Needless to say, we expected the cards to draw rectangles faster than polygons.

Apparently, AMD treats rectangles just like any other polygon, whereas Nvidia employs a more efficient implementation, yielding a benchmark win. Then again, pure rectangles aren’t used all that often, so this is arguably not a big shortcoming. Unfilled objects are usually just drawn as polygons anyway, and this function doesn’t allow the output of rotated rectangles.

Circles, Circle Segments and Ellipses

All cards demonstrate comparable performance in the Arcs and Ellipses benchmark.

Bit Blitting

Bit blitting, which is copying a block from system to graphics memory, is becoming less important. After all, it's the graphics card itself that's supposed to fill its RAM with pixels, not the CPU. Not surprisingly, the performance of this operation hasn't increased much over the past few years. In fact, it actually went in the other direction. Today’s high-end graphics cards barely manage to beat the latest integrated graphics chipsets from Nvidia with full 2D functionality (such as the nForce 630i with GeForce 7100 graphics). Nvidia seems to address the operation a bit better, though in truth every card posts somewhat disappointing results.

Stretching

Stretching is even worse, since the CPU has to help out. Overhead due to the driver getting lost in emulations and offloading computations to the CPU packs quite a whammy.

Summary

Neither Nvidia nor AMD earn a definitive win when it comes to GDI performance. Disappointingly, 2D alacrity seems to be at a standstill, and it has since 2010. At least we've seen AMD alleviate some bottlenecks since then.

In general, applications achieve better performance if they render everything into a temporary DIB (device-independent bitmap) and copy the final result to the graphics card. But at higher screen resolutions, the amount of data that needs to be moved across the PCIe interface can be quite substantial. It's appalling that, in the age of PCIe 3.0, copying data to the graphics card is still an order of magnitude slower than similar operations within the workstation’s RAM.

With that said, AMD's cards perform better than Nvidia's, mostly due to the faster line, spline, and polygon functions. We'd like to think our scathing criticism of AMD’s drivers in the past is partly responsible for this improvement.

The new FirePro W9100 renders complex 2D drawings via GDI almost twice as fast as Nvidia's Quadro K5000. Or only half as slowly. It depends on your point of view.

7. SPECviewperf 12: CATIA, Creo and Maya 2013

Introduction to SPECviewperf 12

SPECviewperf 11, introduced back in 2010, has been showing its age for a while. It wasn't really giving us a realistic-looking picture of modern workstation graphics hardware and driver performance anymore. The applications composing it were just too old. Moreover, AMD and Nvidia were thoroughly optimizing for the specific workloads, throwing off the suite's value.

So, the Standard Performance Evaluation Corporation (SPEC) chose to step up its game with a much-needed update. After all, SPEC’s mission is to create relevant benchmarks that closely adhere to current industry standards.

AMD and Nvidia are both members of SPEC, allowing them to exert some influence over the new collection of tests. The idea is that no company gets an unfair advantage. We'll see how that works out in practice, though.

CATIA V6 R2012

The FirePro W8100 easily surpasses the W8000 that came before it, along with the W9000 and Nvidia Quadro K5000. Of course, Nvidia's Quadro K6000 is in a class of its own. Still, the W8100 demonstrates an impressive debut, landed just behind the W9100.

Creo2

While the new AMD cards top their predecessors, both Quadro cards outperform them. Clearly, AMD’s driver team still has unresolved action items.

Maya 2013

The opposite is true in Maya 2013, where AMD's close collaboration with Autodesk pays off. The FirePro W8100 falls in line behind the W9100, which beats Nvidia's Quadro K6000.

8. SPECviewperf 12: Showcase, Siemens NX and SolidWorks

Showcase 2013

AMD's FirePro W8100 clearly beats the older FirePro W9000 and W8000. Naturally, it can't touch the W9100 though, and it trails Nvidia's first-place Quadro K6000. Still, the Hawaii-powered board is twice as fast as the K5000 based on GK104.

Siemens NX 8.0

Even though the FirePro W8100 lands behind AMD's W9100 (by roughly the same distance we saw in the previous chart), it does manage to pull ahead of AMD's previous-generation professional products. Its predecessor, the W8000, barely outperforms the Quadro K5000. But FirePro W8100 wrecks them both.

SolidWorks 2013 SP1

AMD's FirePro W8100 comes in just behind the W9000, which, in turn, loses to the flagship W9100.

Unfortunately, that finish isn't enough to keep up with Nvidia's top Quadro cards. The K6000 and K5000 finish well ahead of their competition. As you can imagine, then, the performance difference between the FirePro W9000 and Quadro K5000 is humbling, yielding another situation where AMD’s driver team has their work cut out for them.

9. SPECviewperf 12: Synthetic Simulations

Synthetic Tests: Energy

This benchmark simulates a typical volume rendering application, which is used for geophysical surveys (think seismology, along with oil and natural gas exploration) and medical imaging. During the surveys, 2D images are combined to form volumetric representations, creating 2D and 3D views that can be further analyzed and evaluated.

The energy-01 viewset takes advantage of hardware support for 3D textures and the associated trilinear interpolation, which, in turn, depends on a lot of fast graphics memory.

It’s surprising how far AMD’s FirePro workstation graphics cards lag behind Nvidia’s Quadro K6000. The new FirePro W8100 is noticeably beaten by the W9100 and W9000, though it fares well enough to best the FirePro W8000 and Quadro K5000.

Synthetic Tests: Medical

As with the Energy viewset, which covered geophysical surveys and imaging, SPECviewperf 12 uses a synthetic suite to represent the medical field, making use of functionality that is often used for this kind of texture-based volume rendering. Two-dimensional images, created through the use of computer tomography (CT) or magnetic resonance imaging (MRI), are combined into a 3D representation.

The direct volume rendering is achieved by lining up the image slices in parallel. This is done based on texture coordinates, which are specified at every single vertex. They consist of the location in the 3D space (x, y, and z) and also define the alignment and scaling of the texture on the polygon via an object. Next, the values needed for the actual display are calculated based on the texture coordinates. This is called compositing. The entire volume can be thought of as a large number of voxels, or volume pixels, which contain opacity and color on top of the texture information.

Volume ray casting is used to calculate the actual image from the voxels. The present benchmark has two parts. The “4D Heart Data Set” contains several 3D objects, and the “Stag Beetle” places large demands on memory. Comparing AMD’s FirePro W9100 and W8100 using this benchmark shows how cuts to the lower-end board's specs come back to haunt it. Then again, you could also say the W8100's performance is exemplary since it beats the $3200 FirePro W9000.

10. OpenCL: 4K Video Post-Processing

Video Editing and Encoding

For multimedia and entertainment applications, professionals want smooth and efficient processing of high-resolution content. OpenCL and CUDA are well-suited for speeding up such complex calculations.

Because 4K (3840x2160) is becoming more and more common in the professional and desktop spaces, we picked two applications that employ OpenCL to accelerate processing (filtering) and encoding of this up-and-coming format.

We modified our test setup slightly by adding a third SSD, Samsung's 500 GB 840 EVO. It receives the output data, which are large H.264-encoded video files. The input files (several 4K TIFF files and a 4K video) reside on a 480 GB Corsair Neutron GX. We wanted to make sure that storage wasn't introducing any performance-altering bottlenecks.

Adobe Premiere CC Pro

Our two tests include a sequence of TIFF-based images affected by OpenCL-accelerated filters and a high-res video run through another series of filters. This was designed to represent a usage scenario similar to those found in the real world. Filter selection needs to be flexible when doing professional work, after all. AMD's FirePro W8100 trails the W9100 by about 13 percent, which isn’t bad given the ~$800 price difference.

We take a step away from the ultra-demanding workloads suggested in reviewer's guides, strip off the OpenCL-based components, and get down to a more typical benchmark. Under these conditions, the FirePro W8100 pulls inside of the W9000 and barely slips past. Nvidia's Quadro K5000 isn't even a contender here.

Considering that our videos are relatively short, but do include a number of effects, we’re looking at a respectable number of operations that need to be completed. They certainly push our entire system to its limits.

Sony Vegas Pro

This is a good benchmark for AMD's FirePro W8100. The only graphics card it loses to is the pricier W9100.

Overall, the FirePro W8100 does well when you consider its $2500 price point. The advantages of a GPU-accelerated workflow become more pronounced as your content gets longer and the editing more intensive. Workstation-class graphics aren't necessary for manipulating small clips optimized for mobile devices, since the performance differences are barely measurable for that kind of source material.

11. OpenCL: Rendering Performance

LuxMark vs. RatGPU

Meet two different rendering engines that take different approaches. First, there's the popular LuxRender, on which LuxMark is based. This one finally attracted Nvidia's attention after showing up time and again as a weak spot for the company's GeForce and Quadro cards. RatGPU, on the other hand, didn't need that special attention; Nvidia's offerings did well in it right out of the gate.

LuxRender demonstrates that Nvidia's cards do support OpenCL fairly well, if there's no CUDA option. AMD once enjoyed a significant performance advantage in this test, though the magnitude of its wins is shrinking. The following charts represent LuxMark at three difficulty settings:

The FirePro cards land in order of their shader performance for simple single-precision tasks. This changes as the workloads get more complex, allowing the FirePro W8100 to draw even with Nvidia’s Quadro K6000.

Conversely, AMD’s graphics cards don’t do as well in ratGPU. This benchmark isn’t one that gets much attention. Consequently, the two large graphics card vendors don't appear to optimize for it.

Regardless, the rendering approach seems to favor Nvidia’s cards. We once again choose three different difficulty levels.

12. DirectX 11 Gaming: 1920x1080

DirectX11: Gaming at 1920x1080

We’re using data from our 2014 VGA Charts for the normalized performance comparisons. This gives us the best possible overview of whichever benchmarks we want to include. If you would rather look at individual and more detailed results, you'll find them in the charts database. That online comparison tool also lets you pick the boards you'd like to pit against each other in a generated table.

First, let's take a look at 1920x1080 at the highest possible detail settings. Compensating for clock rate differences, the FirePro W8100 would end up at about the same performance level as AMD's Radeon R9 290 in quiet mode. Additional GDDR5 memory doesn't help the workstation card at such a low resolution. That means the FirePro's 17% deficit compared to the desktop board is imposed by the 15% difference between their GPU frequencies.

13. DirectX 11 Gaming: 3840x2160

DirectX: Gaming at 3840x2160

At Ultra HD, it's difficult to push playable frame rates using one graphics card (particularly if the most taxing quality settings are important to you). More interesting to us is how the FirePro W8100 stacks up against the rest of the field.

The Quadro K6000's 18-percent lead over AMD's FirePro W8100 is cut in half to nine percent, and the new Hawaii-based board gains on the Radeon R9 290 as well. It almost catches up to the GeForce GTX 780 Ti 3 GB as well, cutting the consumer card’s 14-percent lead down to only two percent.

Overall, the FirePro's position in this field is acceptable, seeing that gaming is not the point of a workstation graphics card anyway.

14. How We Test Power Consumption

Test Equipment and Test Procedure

Our power consumption test setup was planned in cooperation with HAMEG (Rohde & Schwarz) to yield accurate measurements at small sampling intervals, and we've improved the gear continuously over the past few months.

AMD’s PowerTune and Nvidia’s GPU Boost technologies introduce significant changes to loading, requiring professional measurement and testing technology if you want accurate results. With this in mind, we're complementing our regular numbers with a series of benchmarks using an extraordinarily short range of 100 μs, with a 1 μs sampling rate. We get this accuracy from a 500 MHz digital storage oscilloscope (HAMEG HMO 3054), while measuring currents and voltages with the convenience of a remote control.

The measurements are captured by three high-resolution current probes (HAMEG HZ050), not only through a riser card for the 3.3 and 12 V rails (which was custom-built to fit our needs, supports PCIe 3.0, and offers short signal paths), but also directly from specially-modified auxiliary power cables.

Voltages are measured from a power supply with a single +12 V rail. We're using a 2 ms resolution for the standard readings, which is granular enough to reflect changes from PowerTune and GPU Boost. Because this yields so much raw data, though, we keep the range limited to two minutes per chart.

Measurement Procedure
Contact-free DC measurement at PCIe slot (using a riser card)
Contact-free DC measurement at external auxiliary power supply cable
Voltage measurement at power supply
Measurement Equipment
1 x HAMEG HMO 3054, 500 MHz digital multi-channel oscilloscope
3 x HAMEG HZO50 current probes (1 mA - 30 A, 100 kHz, DC)
4 x HAMEG HZ355 (10:1 probes, 500 MHz)
1 x HAMEG HMC 8012 digital multimeter with storage function


A Lot Can Happen in 100 Milliseconds...

...and we mean a lot! Let’s take a look at an analysis of all three voltage rails using a 2 ms sample across 100 ms (giving us 50 data points). Looking at those results makes us pity the power supply.

The information we're collecting is interesting enough that we make full use of the oscilloscope’s storage to achieve an even more exact picture of each card's real power consumption.

15. Power Consumption: Detailed Results

Power Consumption: Idle

Since workstation graphics cards are often in-use more than not, the consumption floor isn't as important to us. Still, though, we all want to know where the FirePro idles.

At 14 W, the W8100 lands a bit higher than AMD's Radeon R9 290, and the extra GDDR5 isn't enough to explain the delta. My best guess would be that a beefier power supply and a higher minimum voltage push our measurement of the professional board higher. The approximately 1.8 W of additional draw at idle is nothing to worry about, though.

Power Consumption Scenario: Primarily Graphics

Rendering and compute workloads can’t really be teased apart completely, since so many of today's tasks are mixed. The FirePro W8100’s 188.4 W measurement is good, especially since it almost matches the Quadro K6000’s 188.7 W.

Those peaks above 300 W shouldn’t pose any major problems for decent PSUs. The less-than-70 W maximum load demanded from the motherboard isn’t an issue, either.

Power Consumption Scenario: Primarily Computing

A look at the curve illustrates that peaks are a lot less pronounced when more constant compute-oriented tasks are executed. They can't be avoided entirely, though.

It’s a lot more interesting that the measured power consumption is almost exactly the same as in the previous chart. As a result, we now know that 190 W is the boundary AMD's FirePro W8100 just won’t cross.

In comparison, Nvidia's Quadro K6000 registers 202.3 W in this scenario, which means that it draws almost 12 W more than the FirePro W8100. Depending on the application, AMD’s new workstation card can compete a bit better in benchmarks of efficiency. Catching up or even passing Nvidia's flagship in this discipline is out of the question. Nevertheless, what AMD accomplishes with its latest offering is a positive development on many levels.

16. Heat and Noise

Temperatures

Beyond the measurements we take on an open test bench, we also built a fixed workspace that has one purpose: testing graphics cards under more realistic conditions using a closed case. Corsair's Graphite 760T by HEC/Compucase jumped out at us immediately because of its acrylic side panel that can be opened like a door.

We run the case fans at maximum speed, since lots of cooling is normal in the server and workstation spaces. Still, a closed chassis is far from an ideal environment. Or is it? Interestingly, an open or closed side door has very little impact on graphics cards employing direct heat exhaust.

Let’s first compare the curves from AMD's FirePro W9100 and W8100. The flagship board reaches 93 degrees Celsius, regardless of whether the enclosure is open or closed, throttling back as a result of overly-aggressive factory settings. Conversely, the W8100 maintains a ceiling about six degrees cooler thanks to its lower power consumption. It too throttles, though. PowerTune has to step in and help starting at around 77 degrees Celsius.

As we'd deduce, based on those observations, the W8100's fan spins slower, which means the card should make less noise.

Noise

We measure each graphics card's acoustic behavior with a calibrated high-quality studio microphone (supercardioid) 50 cm away from a position perpendicular to the middle of the board. This distance, as well as the strong cardioid microphone characteristic, represents a compromise between avoiding noise generated by the fan’s airflow and ambient noise that can never be completely eliminated. Our dampening efforts help minimize the latter, but they'll never be 100-percent successful.

As we've seen many times before, reference-class cards typically achieve their cooling performance at the cost of higher sound levels. High-end workstation cards, in particular, exhaust waste heat from their I/O panels to avoid affecting other platform components. However, this is enabled through the use of a radial fan, and our results show that they're quite noisy.

Here are the detailed results:

Model
Idle
3D Workload, Open Lab Table
3D Workload, Closed Case
Quadro K5000
30.8 dB(A)
37.7 dB(A)
37.1 dB(A)
Quadro K6000
30.8 dB(A)
42.7 dB(A)
41.2 dB(A)
FirePro W9100
33.5 dB(A)51.3 dB(A)
49.8 dB(A)
FirePro W8100
32.3 dB(A)
44.8 dB(A)
43.5 dB(A)
FirePro W9000
33.2 dB(A)55.4 dB(A)52.7 dB(A)

Video

The video shows that the AMD FirePro W8100 is bearable when it comes to maximum noise under load. This also demonstrates that a thermal solution originally designed for the Radeon HD 5800 (which hasn't changed much since) deals with the W8100’s nearly 190 W a lot better than the W9100's 250 W.

AMD FirePro W8100 - GPGPU - 96 to 100 Percent Load

17. A Jack Of All Trades For A Good Price

AMD is trying to close a gap in its line-up between the compute performance-oriented FirePro W9100 and the highly capable W7000 with the new FirePro W8100. It's a jack of all trades, so long as your applications are optimized for what it can do, and that $2500 price point doesn't scare you away.

A Combination of Strong Rendering And Compute Performance Yields A Well-Rounded Card

Due to its design and purpose, AMD's FirePro W8100 is perhaps the most well-rounded solution for tasks heavy in 3D rendering and and complex computing, such as CAD, CAE, multimedia, and entertainment. It pushes the $800-pricier FirePro W9100 out of the way in the same way AMD's Radeon R9 290 outshines the 290X for its superior value.

In that way, if you skipped over the FirePro W9100 due to its cost, you might be tempted by AMD’s newest offering. It has to be said that the W8100's purpose, defined on the first page of this story, is achieved perfectly, resulting in a product that’s hard for the performance-oriented professional to resist.

As long as AMD enjoys continued success promoting OpenCL support in workstation-class applications, the company's FirePro family should continue regaining market share. Officially embracing the Mac platform is a step in the right direction, even though the volume of Mac-compatible cards is still low.

4K Resolution and Connectivity Galore

The FirePro W8100 can drive up to four 4K monitors at 30 Hz, or three at 60 Hz. Its six display engines make it possible to attach that many screens through MST hubs, too. Only the FirePro W9100 offers a more advanced connectivity suite by supporting as many as six 4K screens at 30 Hz. 

Eight gigabytes of on-board GDDR5 memory is sufficiently large; no application we tested was able to push that capacity to its limit. In this price range, then, AMD has no real competition. Nvidia's Quadro K5000 is quite a bit cheaper, but comes with half the memory, less 4K connectivity, and is generally slower. It takes a Quadro K6000 to beat it, and that board sells for $5000.

Cooling and Power Consumption

One opportunity for improvement is AMD's underwhelming thermal solution, which we've seen previously on the company's desktop-oriented reference cards. By redesigning the cooler, some board partners have already demonstrated that Hawaii can be made to run at much lower temperatures than 92 or 87 degrees Celsius. The challenge, of course, is that those gaming products start exhausting heat inside your chassis, and that just doesn't fly in the workstation world.

Instead, professional cards need to push thermal energy out from their I/O brackets. Nvidia does this successfully with its Quadro cards, and AMD should start following suit. The FirePro W8100's heat sink and fan undoubtedly sacrifice some of the board's performance potential, since Hawaii is known to perform best under optimal cooling.

Bottom Line

In light of its price, the AMD FirePro W8100 is a compelling piece of hardware. It’s well-rounded and, consequently, fares well in a variety of applications. It also has a large number of connectivity options for high-resolution displays. The balanced mix of good graphics performance and a strong showing in our OpenCL-accelerated tests makes the W8100 particularly attractive in environments where both are needed, either for a specific application or the parallel usage of different applications.

It’s also nice to see that AMD closed the gap between its FirePro W7000 and flagship W9100 without making major cuts to performance. The W8100 can replace AMD’s former flagship, the W9000, in almost all areas, and boasts a lot of additional features to boot. It’ll be interesting to see if the company introduces another workstation graphics card under the W8100, maybe at a $2000 price point, when the Tonga chips are introduced.