XGI Volari 8300 128 MB 64-bit Review

[nextpage title=”Introduction”]

Volari 8300 is the newest entry-level graphics chip from XGI. It is always good to see another company besides ATI and NVIDIA doing something. The reference board we’ve got from XGI had 128 MB on-board memory and a 64-bit memory interface, competing with Radeon X300 SE HyperMemory and with the 64-bit 64 MB version of GeForce 6200 TurboCache. XGI’s chip, however, has a built-in MPEG-2 decoder and video enhancer, i.e., with this video card DVD decoding is done by the video card and not by the system CPU, which can be a tremendous feature for entry-level multimedia PCs. GeForce 6200 also has this feature, but you need to buy a decoder software from NVIDIA that costs between USD 19.95 and USD 49.95, depending on the version (you can use it for free during the first 30 days). Radeon X300 doesn’t have this feature at all, while the new Radeon X1300 series has it.

If you are interested in knowing more about this subject, please read our tutorial Enabling 2D Enhancements on GeForce Series 6 and 7.

Figure 1: Volari 8300 from XGI.

Volari 8300 runs at 300 MHz with its memory running at 600 MHz using a 64-bit interface, just like Radeon X300 SE HyperMemory and some GeForce 6200 TurboCache (there are boards from this series with 32-bit memory interface). It graphics engine is DirectX 9.0, supporting Shader 2.0, so it is similar to Radeon X300, as both GeForce 6200 and Radeon X1300 are based on Shader 3.0, i.e., DirectX 9.0c. In our tests, however, this video card didn’t support antialising. According to XGI, this feature was not implemented because of the market this product is targeted (very entry-level) and to reduce its price.

Talking about pricing, XGI suggested price for this product is only USD 49.95, below the current market price for both Radeon X300 SE HyperMemory and GeForce 6200 TurboCache.

You can see in our tutorial “NVIDIA Chips Comparison Table” the difference between Volari 8300 chip and its competitors from NVIDIA, while on our tutorial “ATI Chips Comparison Table” you can compare it to its competitors from ATI.

Let’s now take a closer look at the reference board for Volari 8300 from XGI.

[nextpage title=”The Volari 8300 Reference Board from XGI”]

On Figures 2 and 3 you can check Volari 8300 reference board from XGI. As you can see, this video card uses an active heatsink (i.e., with a fan).

Figure 2: XGI Volari 8300.

Figure 3: XGI Volari 8300, back view.

We removed the heatsink to take a look at the memory chip, and as you can see in Figure 4, this card uses two 512-Mbit memory chip.

Figure 4: The reviewed card without its heatsink.

This video card uses two DDR 512-Mbit 2.8 ns chips from Hynix (HHY5DS113222FM-28)to give it 128 MB of video memory (512 Mbits x 2 = 128 MB). These chips can run up to 700 MHz. Since this video card accesses the memory at 600 MHz there is a 17% headroom for memory overclocking inside the memory’s specifications. The problem, however, is finding a tool to overclock it, since PowerStrip software does not recognize this chip yet and the video driver doesn’t have any overclocking option.

Figure 5: 2.8 ns DDR memory chip used by XGI Volari 8300.

[nextpage title=”Main Specifications”]

• Graphics chip: Volari 8300 running at 300 MHz.
• Memory: 2.8 ns 64-bit 128 MB DDR memory from Hynix (HHY5DS113222FM-28), running at 600 MHz.
• Bus type: PCI Express 16x.
• Connectors: One VGA, one DVI and one mini-DIN for S-Video output.
• Number of CDs that come with this board: N/A.
• Games that come with this board: None.
• Programs that come with this board: None.
• More Information: https://www.xgitech.com
• Suggested Price: USD 49.95

[nextpage title=”How We Tested”]

During our benchmarking sessions, we used the configuration listed below. Between our benchmarking sessions the only variable was the video card being tested.

Hardware Configuration

• Motherboard: Intel D915GEV
• CPU: Pentium 4 3.4 GHz LGA 775
• Memory: Two 512 MB DDR2-533 CM2X512-4200 CL4 from Corsair memory modules
• Hard Drive: Maxtor DiamondMax 9 Plus (40 GB, ATA-133)
• Screen resolution: [email protected] Hz

Software Configuration

• Windows XP Professional installed using NTFS
• Service Pack 2
• Direct X 9.0c
• Intel inf driver version: 7.2.2.1006
• ATI video driver version: 5.11
• NVIDIA video driver version: 81.95
• Intel video driver version: 14.17
• XGI video driver version: 3.01.130.D (6.14.1.3010)

Used Software

• 3DMark2001 SE
• 3DMark03 Business 3.60
• 3DMark05 Business 1.20
• Doom 3 1.3
• Far Cry 1.33

We adopted a 3% error margin; thus, differences below 3% cannot be considered relevant. In other words, products with a performance difference below 3% should be considered as having similar performance.

[nextpage title=”3DMark2001 SE”]

3DMark2001 SE measures video card performance simulating DirectX 8.1 games. It is very effective software for evaluating the performance from previous-generation games, programmed using DirectX 8. In this software we ran two tests, both at 1024x768x32. Since we were evaluating low-end video cards, we decided to not run our tests in higher resolutions, since rarely a user that buys a video card from this level will push resolutions above 1024×768 in 3D games.

We ran this software without antialising and with no frame buffer. Unfortunately Volari 8300 does not support antialising, so we couldn’t compare it to other video cards enabling 3D video quality enhancements.

You may be asking yourself why we added an old program in a review of a latest generation video card. To us, it is as important to know the performance of a video card with the latest games as it is to know its performance in an older game. That’s why we kept this software in our methodology.

At the default 3DMark2001 SE configuration, XGI Volari 8300 128 MB was betean by Radeon X1300 Pro 256 MB 128-bit (ATI), which was 176.58% faster, GeForce 6600 128 MB (Albatron), which was 139.11% faster, GeForce 6600 GT 128 MB (NVIDIA), which was 132.17% faster, GeForce 6200 128 MB 128-bit (Leadtek), which was 122.92% faster, GeForce 6200 TC 64 MB 64-bit (XFX), which was 64.64% faster, Radeon X300 SE HM 128 MB 64-bit (PowerColor), which was 32.42% faster, Radeon X300 128 MB 128-bit (ATI), which was 32.33% faster, % faster,
GeForce 6200 TC 16 MB 64-bit (Leadtek), which was 26.93% faster and i915G (Intel D915GEV), which was 7.82% faster.

[nextpage title=”3DMark03″]

3DMark03 measures performance by simulating games written to DirectX 9, which are contemporary games. In this software we ran two tests, both at 1024x768x32. Since we were evaluating low-end video cards, we decided to not run our tests in higher resolutions, since rarely a user that buys a video card from this level will push resolutions above 1024×768 in 3D games.

We ran this software without antialising and with no anisotropic filtering. Unfortunately Volari 8300 does not support antialising, so we couldn’t compare it to other video cards enabling 3D video quality enhancements.

At the default 3DMark03 configuration, XGI Volari 8300 128 MB was 8.79% faster than GeForce 6200 TC 16 MB 64-bit (Leadtek), 19.95% faster than Radeon X300 128 MB 128-bit (ATI), 19.95% faster than Radeon X300 SE HM 128 MB 64-bit (PowerColor) and 54.11% faster than i915G (Intel D915GEV).

XGI Volari 8300 128 MB was beaten by GeForce 6600 GT 128 MB (NVIDIA), which was 273.06% faster, Radeon X1300 Pro 256 MB 128-bit (ATI), which was 160.38% faster, GeForce 6600 128 MB (Albatron), which was 129.91% faster, GeForce 6200 128 MB 128-bit (Leadtek), which was 76.97% faster and GeForce 6200 TC 64 MB 64-bit (XFX), which was 13.41% faster.

[nextpage title=”3DMark05″]

3DMark05 measures performance by simulating DirectX 9.0c games, i.e., using the new Shader 3.0 model. This programming model is used by Far Cry game and other games to be launched in the future. This new programming model is used by GeForce 6 and 7 series from NVIDIA and Radeon X1000 series from ATI.

In this software we ran two tests, both at 1024x768x32. Since we were evaluating low-end video cards, we decided to not run our tests in higher resolutions, since rarely a user that buys a video card from this level will push resolutions above 1024×768 in 3D games.

We ran this software without antialising and with no anisotropic filtering. Unfortunately Volari 8300 does not support antialising, so we couldn’t compare it to other video cards enabling 3D video quality enhancements.

At the default 3DMark05 configuration, XGI Volari 8300 128 MB was 3.70% faster than GeForce 6200 TC 16 MB 64-bit (Leadtek) and 191.91% faster than i915G (Intel D915GEV).

XGI Volari 8300 128 MB was beaten by GeForce 6600 GT 128 MB (NVIDIA), which was 254.85% faster, Radeon X1300 Pro 256 MB 128-bit (ATI), which was 177.82% faster, GeForce 6600 128 MB (Albatron), which was 108.42% faster, GeForce 6200 128 MB 128-bit (Leadtek), which was 52.67% faster, GeForce 6200 TC 64 MB 64-bit (XFX), which was 31.78% faster, % faster, Radeon X300 128 MB 128-bit (ATI), which was 10.30% faster and Radeon X300 SE HM 128 MB 64-bit (PowerColor), which was 9.90% faster.

[nextpage title=”Doom 3″]Doom 3 is one of the heaviest games available today. As we’ve done on other programs, we ran this game at three resolutions: 1024x768x32, 1280x1024x32 and 1600x1200x32. This game allows several image quality levels and we’ve done our benchmarking on two levels, low and high. We ran demo1 four times and wrote the obtained number of frames per second. The first result we discarded at once, since it is far inferior than the other results. This happens because at the first time we run the demo the game must load all textures to video memory, fact that doesn’t happen from the second time we run the demo on. From the three results left, we consider as our official result the middle result, i.e., we discard the highest and the lowest values. Curiously almost all times the values obtained at the second round on were the same.

A very important detail that we must mention is that Doom 3 has an internal FPS lock: it is only capable of generating 60 frames per second, even if your board is able to produce more frames per second than that. This is done in order to make the game to have the same “playability” sensation independently from the video card installed on the PC. This lock, however, is disabled in the game benchmarking mode.

For further details on how to measure 3D performance with Doom 3, read our tutorial on this subject.

Running this game in its low video quality mode, XGI Volari 8300 128 MB was beaten by GeForce 6600 GT 128 MB (NVIDIA), which was 715.56% faster, GeForce 6600 128 MB (Albatron), which was 521.11% faster, Radeon X1300 Pro 256 MB 128-bit (ATI), which was 402.22% faster, GeForce 6200 128 MB 128-bit (Leadtek), which was 380.00% faster, GeForce 6200 TC 64 MB 64-bit (XFX), which was 121.11% faster, GeForce 6200 TC 16 MB 64-bit (Leadtek), which was 64.44% faster, % faster, Radeon X300 SE HM 128 MB 64-bit (PowerColor), which was 55.56% faster and Radeon X300 128 MB 128-bit (ATI), which was 55.56% faster.

Enabling video quality enhancements, XGI Volari 8300 128 MB was beaten by GeForce 6600 GT 128 MB (NVIDIA), which was 846.05% faster, GeForce 6600 128 MB (Albatron), which was 589.47% faster, Radeon X1300 Pro 256 MB 128-bit (ATI), which was 453.95% faster, GeForce 6200 128 MB 128-bit (Leadtek), which was 431.58% faster, GeForce 6200 TC 64 MB 64-bit (XFX), which was 143.42% faster, GeForce 6200 TC 16 MB 64-bit (Leadtek), which was 85.53% faster, % faster, Radeon X300 SE HM 128 MB 64-bit (PowerColor), which was 76.32% faster and Radeon X300 128 MB 128-bit (ATI), which was 76.32% faster.

[nextpage title=”Far Cry”]

Far Cry is a game based on the new Shader 3.0 (DirectX 9.0c) model, which is used by GeForce 6 and 7 series from NVIDIA and Radeon X1000 series from ATI.
 
As we’ve done on other programs, we ran this game only at 1024×768. Since we were evaluating low-end video cards, we decided to not run our tests in higher resolutions, since rarely a user that buys a video card from this level will push resolutions above 1024×768 in 3D games.

This game allows several image quality levels and we’ve done our benchmarking only on low level, since Volari 8300 doesn’t support antialising. To measure the performance we used the demo created by German magazine PC Games Hardware (PCGH), available at https://www.3dcenter.org/downloads/farcry-pcgh-vga.php. We ran this demo four times and made an arithmetical average with the obtained results. This average is the result presented in our graphs.

For further details on how to measure 3D performance with Far Cry, read our tutorial on this subject.

Running this game in its low video quality mode, XGI Volari 8300 128 MB was beaten by Radeon X1300 Pro 256 MB 128-bit (ATI), which was 236.40% faster, GeForce 6600 GT 128 MB (NVIDIA), which was 220.80% faster, GeForce 6600 128 MB (Albatron), which was 211.55% faster, GeForce 6200 128 MB 128-bit (Leadtek), which was 146.70% faster, GeForce 6200 TC 64 MB 64-bit (XFX), which was 58.00% faster, Radeon X300 SE HM 128 MB 64-bit (PowerColor), which was 48.00% faster, Radeon X300 128 MB 128-bit (ATI), which was 46.95% faster and GeForce 6200 TC 16 MB 64-bit (Leadtek), which was 21.53% faster.

[nextpage title=”Conclusions”]

According to XGI, Volari 8300 is targeted to multimedia PCs, and does not try to bring a great 3D performance. It is cheaper than its main competitors and yet brings a good performance on 3DMark03 (if we think about its price, of course) and a very interesting 3D performance on 3DMark05.

It has hardware-based DVD decoding. This is a great feature for entry-level PCs, because on computers without this feature it is the system CPU that decodes the DVD. If you don’t have a very powerful processor – as it is almost always the case when we think about very entry-level PCs – the CPU will be overloaded and will start dropping frames during DVD decoding. In other words, the CPU will be too busy that it will not decode everything, thus dropping video quality.

In our tests, with DVD decoding done by software (i.e., by the system CPU), CPU load was between 12% and 15%. Enabling DVD decoding to be done by hardware (i.e., by the video card), CPU load dropped to below 4%. We were using a 3.4 GHz Pentium 4 and we could see this big difference in performance; in a more modest system the difference can be huge.

Also, keep in mind that if you have a GeForce 6200 you need to buy an extra piece of software to enable its DVD decoding that costs between USD 19.95 and USD 49.95, depending on the version (amazingly enough this same software was priced at only USD 15 not six months ago). Radeon X300 does not provide this feature. Radeon X1300 does, but this video card is quoted at the USD 100 range (for the 256 MB version, we couldn’t find the 128 MB for sale yet), double the price of Volari 8300.

So, Volari 8300 provides an honest 3D performance and a great DVD decoding done by hardware, meaning that you will have a terrific DVD playback even if you system has a very entry-level CPU. We think that this is a great product for multimedia PCs, and we think that more and more people are buying PCs to watch videos, download pictures from their digital cameras and to listen to MP3 files. For this purpose, we think Volari 8300 price is just right, and as far as we know it is the cheapest PCI Express video card available today.

We wish to congratulate XGI for finally understanding that pricing is everything on this market. Volari V8 chip, for example, provided a good mid-range performance when it was released, but it was quoted at the same price level of competing chips from ATI and NVIDIA. It is very hard for a newcomer to compete head-to-head with the leaders, even if you have the best product on the market. You should drop the price to gain market share. It seems that XGI has finally realized that.

The success of this product will depend a lot on XGI ability to deliver the message that this product is not intended to be a 3D gaming machine (even though it provides a reasonable 3D performance for its low price), and that what is really awesome on it is its DVD decoding feature, and that someone looking for this feature will pay half the price on a Volari 8300 than on a GeForce 6200 + DVD decoder software or on a Radeon X1300.

We must also remember that it is not a matter of installing this video card to have its hardware-based DVD decoding. You should configure your DVD player software to use this feature, which is not enabled by default. Let’s hope XGI can make this very clear on its final product.

XGI’s Achilles Heel is availability. In the past they tried to find partners to manufacture video boards using their chips, but all video card manufacturers were too committed with NVIDIA and/or ATI and didn’t want to risk building XGI-based VGAs. Now XGI is selling their products under their own brand, let’s hope they have luck on distributing them.