Be sure to read Part 1 of our KX133 Review for more information on the chipset.

Only a couple of weeks ago, we took a look at the performance of VIA’s recently released KX133 chipset for the Athlon.  Since then, we have had the opportunity to take a look at a couple of shipping motherboards based on the KX133 chipset and have decided to provide an update on performance scores using shipping boards as well as performance benchmarks under Windows NT.

From the manufacturers we have talked to, their KX133 based motherboards are essentially complete, with the only thing left to be addressed being BIOS tweaking and any compatibility issues that may pop up.  It is partially because of this that we are revisiting the chipset to see how the shipping performance of the KX133 compares to the reference board that we benchmarked not too long ago because, after all, you can’t go out and purchase VIA’s reference design with their extremely configurable BIOS setup to tweak your performance. 

One of the areas of performance that didn’t make it into Part 1 of our KX133 review was professional level OpenGL performance.  However, since then, we have received updated Windows NT 4.0 drivers (version 3.76) that dramatically improve memory and AGP performance under Windows NT with the KX133 chipset. 

We asked NVIDIA exactly what these drivers enabled with the KX133 and this was their explanation:

Please find attached an un-numbered driver which fully supports AGP 4X in VIA's KX-133 Chipset for Athlon under Windows NT4. These modifications to our existing driver were made as soon as we received the preliminary board that VIA sent us last week and will be included in a driver release when motherboards using VIA's KX-133 are readily available for purchase. This Chipset was not available for verification of the Chipset ID at the time that the 3.68 driver was released and AGP support is not generically enabled by Windows NT4 for any devices.

It is written specifically for each separate component by our Graphics Driver team once we have the appropriate silicon in hand. Every time a new Core Logic component is released to the market from AMD, Intel etc... the NVIDIA Driver team adds AGP support for that device to our NT4 Graphics driver shortly after we receive the product.

AMD received a driver similar to this one last Friday(1/28) to do some preliminary testing, but, the driver below has been created uniquely for your testing and has not been passed through qualification testing. Please keep the driver id and number confidential and do not release this driver outside of your testing labs. It is not intended for general usage.

EPoX 7KXA Click to Enlarge

Regardless of what the driver actually enables/disables, the fact of the matter is it significantly improves the professional level OpenGL/NT performance of NVIDIA products on the KX133 platform.

Whether or not this was actually a conspiracy to hide the Athlon’s true high-end performance is up to you to decide, but to us it seems like a case of NVIDIA not having a final KX133 sample in time for the improvements to be included in their last driver release. 

Armed with EPoX's newly shipping 7KXA KX133 motherboard, we took the KX133 back to the tests to make sure that the shipping performance of the chipset was up to par with what VIA's reference design provided us with not too long ago.

 

The Test

The performance difference between the three Athlon chipsets is negligible under Content Creation Winstone 2000. Excluding the lower outlier (VIA 133A), the five other test systems perform virtually identically to one another. The difference in performance can be attributed to normal variations in test scores and isn't significant, with the exception of the 29.2 VIA 133A reference score.

 

We have a similar situation under SYSMark 2000, the performance difference between the EPoX 7KXA test bed using the KX133 chipset and the Gigabyte GA-7IX (AMD 750) with SuperBypass enabled is negligible The real performance advantage here is of both chipsets over the initial steppings of the AMD 750 chipset, more specifically the AMD 751 North Bridge with no support for SuperBypass.

The EPoX 7KXA (KX133) seemed to keep up quite well with our Gigabyte GA-7IX (AMD 750) test bed, but distanced itself quite nicely from the 7IX with SuperBypass disabled. The 0.6 fps performance difference between the KX133 and AMD 750 - SuperBypass test setups is negligible, once again the biggest improvement is that held over the original steppings of the 750 without SuperBypass support.

At 1024 x 768 x 32 the DDR GeForce is fillrate limited and thus all of the scores hover within the 61 - 62 fps range.

Here is where the KX133 truly begins to shine. UnrealTournament is very texture dependent and thus benefits greatly from the faster memory bus of the KX133. The performance improvement here is quite hefty at a full 22%. The KX133 with the 133MHz memory bus frequency helps to bridge the performance gap between the Athlon and the Pentium III platform here.

We get a similar boost at 1024 x 768 x 32, but this time the improvement is "only" 19%. Not bad for a chipset that's cheaper than the AMD 750. Once again, because of the KX133, the Athlon is now much more competitive in situations such as this one.

If we take the i820 test platform out of the picture because of the unreasonable price of RDRAM, the Athlon on the KX133 platform actually ends up being the fastest under Expendable, a benchmark that had previously favored the Pentium III. It should also be noted that the SuperBypass enabled AMD 750 system performed virtually on-par with the KX133, once again illustrating the point that the KX133 isn't too much faster than the AMD 750 with SuperBypass enabled in most consumer level applications, but luckily it's not slower. Combined with the lower overall cost of the KX133 solution, VIA's solution is definitely a viable one.

We have a similar situation at the higher resolutions, the KX133 gives the Athlon a nice boost to just below the Pentium III 800/RDRAM setup. Keep in mind that this on a production level, shipping KX133 motherboard, this is the minimum performance level you should expect from the KX133 and Athlon 800 combo.

SPECviewperf

The Standard Performance Evaluation Corporation, commonly known as SPEC, managed to come up with a synthetic benchmark with real world implications. By running specific "viewsets" SPECviewperf can simulate performance under various applications. To be more accurate, according to SPEC, "A viewset is a group of individual runs of SPECviewperf that attempt to characterize the graphics rendering portion of an ISV's application." While this method is by no means capable of identifying the performance of a card in all situations, it does help to indicate the strengths and weaknesses of a particular setup.

SPECviewperf 6.1.1 currently features five viewsets: the Advanced Visualizer, the DesignReview, the Data Explorer, the Lightscape and the ProCDRS-02 viewset. Before each benchmark set we've provided SPEC's own description of that particular viewset so you can better understand what that particular viewset is measuring, performance-wise.

Each viewset is divided into a number of tests, ranging from 4 to 10 in quantity. These tests each stress a different performance element in the particular application that viewset is attempting to simulate. Since all applications focus on some features more than others, each one of these tests is weighted meaning that each test affects the final score differently, some more than others.

All results are reported in frames per second, so the higher the value, the better the performance is. The last result given for each of the viewsets is the WGM or Weighted Geometric Mean. This value is, as the name implies, the Weighted Geometric Mean of all of the test scores. The formula used to calculate the WGM is as follows:

With n being the number of tests in a viewset and w being the weight of each test expressed as a number between 0.0 and 1.0.

If you'd like to know more about why a Weighted Geometric Mean is used, SPEC has an excellent article detailing just why, here.

We ran the SPECviewperf 6.1.1 package under NT for a high-end workstation performance comparison. A noteworthy change has been made since the last Athlon CPU review, we have started using a DDR GeForce in the high-end tests after discovering that in spite of the GeForce's hardware T&L engine the card uses virtually all of the host CPU during the SPECviewperf tests thus putting a great strain on the CPU.

We also received an updated set of drivers (v3.76) for Windows NT that fully take advantage of the KX133 chipset which help to produce some very interesting benchmark results

Advanced Visualizer (AWadvs-03) Viewset

Taken from http://www.spec.org/gpc/opc.static/awadvs.htm

Advanced Visualizer from Alias/Wavefront is an integrated workstation-based 3D animation system that offers a comprehensive set of tools for 3D modeling, animation, rendering, image composition, and video output. All operations within Advanced Visualizer are performed in immediate mode with double buffered windows. There are four basic modes of operation within Advanced Visualizer:

 
• 55% material shading (textured, z-buffered, backface-culled, 2 local lights)
• 95% perspective, 80% trilinear mipmapped, modulated (41.8%)
• 95% perspective, 20% nearest, modulated (10.45%)
• 5% ortho, 80% trilinear mipmapped, modulated (2.2%)
• 5% ortho, 20% nearest, modulated (.55%)
• 30% wireframe (no z-buffering, no lighting)
• 95% perspective (28.5%)
• 5% ortho (1.5%)
• 10% smooth shading (z-buffered, backface-culled, 2 local lights)
• 95% perspective (9.5%)
• 5% ortho (.5%)
• 5% flat shading (z-buffered, backface-culled, 2 local lights)
• 95% perspective (4.75%)
• 5% ortho (.25%)

SPECviewperf's first test using the Advanced Visualizer (AWadvs-03) viewset simulates the usage of a workstation level 3D animation setup, and the most obvious point is that the Athlon on an AMD 750 chipset, even with SuperBypass enabled is performing 30% slower than an equivalently clocked Pentium III on a BX setup. This is where the KX133 truly shines, in professional level applications that are very memory intensive and take advantage of the AGP 4X capabilities of the chipset for transferring geometry data to the graphics controller.

DesignReview (DRV-06) Viewset

Taken from http://www.spec.org/gpc/opc.static/drv.htm

DesignReview is a 3D computer model review package specifically tailored for plant design models consisting of piping, equipment and structural elements such as I-beams, HVAC ducting, and electrical raceways. It allows flexible viewing and manipulation of the model for helping the design team visually track progress, identify interference, locate components, and facilitate project approvals by presenting clear presentations that technical and non-technical audiences can understand. There are 6 tests specified by the viewset that represent the most common operations performed by DesignReview.

The DesignReview (DRV-06) Viewset is another 3D modeling test and as we saw from the previous test, the AMD 750 is still lagging behind. This time around the KX133 is ever so slightly edged out of first place by the Pentium III 800, but the fact that the KX133 offers a offers close to a 60% performance improvement over the AMD 750 chipset combined with the fact that this is a real world performance comparison with actual shipping motherboards makes the KX133 even more impressive.

The KX133 does help the Athlon remain competitive with the Pentium III in the areas that it was falling behind in, but as we just mentioned the main comparison is between the AMD 750 and the KX133 because, after all, this is a review of an Athlon chipset not a Pentium III chipset.

Data Explorer (DX-05) Viewset

Taken from: http://www.spec.org/gpc/opc.static/dx.htm

The IBM Visualization Data Explorer (DX) is a general-purpose software package for scientific data visualization and analysis. It employs a data-flow driven client-server execution model and is currently available on Unix workstations from Silicon Graphics, IBM, Sun, Hewlett-Packard and Digital Equipment. The OpenGL port of Data Explorer was completed with the recent release of DX 2.1.

The tests visualize a set of particle traces through a vector flow field. The width of each tube represents the magnitude of the velocity vector at that location. Data such as this might result from simulations of fluid flow through a constriction. The object represented contains about 1,000 triangle meshes containing approximately 100 verticies each. This is a medium-sized data set for DX.

The performance throne is restored to the Athlon 800/KX133 setup in the Data Explorer (DX-05) Viewset, and the KX133 vs AMD 750 performance advantage is pushed up to an 80% performance increase provided by the KX133. This type of an improvement is incredible, but part of the performance improvement is because of the 133MHz memory bus frequency, regardless, for professional level OpenGL users, the Athlon/KX133 is an available one and a powerful one.

Lightscape (Light-03) Viewset

Taken from: http://www.spec.org/gpc/opc.static/light.htm

The Lightscape Visualization System from Discreet Logic represents a new generation of computer graphics technology that combines proprietary radiosity algorithms with a physically based lighting interface.

There are four tests specified by the viewset that represent the most common operations performed by the Lightscape Visualization System

Here the KX133 is almost 2.5 times faster than the AMD 750, about 30 - 50% of the performance boost over the AMD 750 is due to the 133MHz memory bus frequency but that still leaves an additional 60 - 80% that is due to the chipset and driver optimizations alone.

ProCDRS-02 Viewset

Taken from: http://www.spec.org/gpc/opc.static/procdrs.htm

The ProCDRS-02 viewset is a complete update of the CDRS-03 viewset. It is intended to model the graphics performance of Parametric Technology Corporation's CDRS industrial design software.

For more information on CDRS, see http://www.ptc.com/icem/products/cdrs/cdrs.htm

The viewset consists of ten tests, each of which represents a different mode of operation within CDRS. Two of the tests use a wireframe model, and the other tests use a shaded model. Each test returns a result in frames per second, and a composite score is calculated as a weighted geometric mean of the individual test results. The tests are weighted to represent the typical proportion of time a user would spend in each mode.

All tests run in display list mode. The wireframe tests use anti-aliased lines, since these are the default in CDRS. The shaded tests use one infinite light and two-sided lighting. The texture is a 512 by 512 pixel 24-bit color image.

Finally we have the ProCDRS-02 Viewset, in which the KX133 comes out on top of the competition including the AMD 750, which it beats by about 50%.

Conclusion

The performance of the KX133 hasn't changed much from our reference board supplied by VIA to the shipping boards we've received, including the EPoX 7KXA which we used for the tests in this comparison. This is a good thing, because in the past, motherboard manufacturers have had problems properly implementing VIA chipsets into their designs, generally because they would simply drop a VIA North Bridge into their existing P6 bus designs with very few modifications outside of the bare minimum requirements. Luckily this isn't physically possible with the KX133 chipset, forcing motherboard manufacturers to produce completely original designs based on the VIA reference layout.

The next release of NVIDIA's Detonator drivers for Windows NT will feature the performance enhancements and optimizations that propelled our KX133 test bed to the top of the charts under the SPECviewperf tests. For those of you that own KX133 setups now and need the performance boost immediately, you can easily find the beta drivers online if you search around a bit.

As more KX133 based boards hit the streets, we can hope that the performance will improve as manufacturers begin to tweak the BIOS files more for speed. The only thing there is to be worried about is if motherboard manufacturers decide that promoting their KX133 products isn't worth the potential risks caused by VIA's competitors. If you remember, this was the case with the earliest AMD 750 based motherboards and their manufacturers, while we are expecting things to change with the KX133, there will still be a few frightened manufacturers, let's hope those frightened few don't become the majority this time around.

Be sure to read Part 1 of our KX133 Review for more information on the chipset.