Telling It Like
It Is Since 1985

Bits & BytesSearch EnginesCrime WatchEmail STReport

App Reviews
Game Reviews
Editor's Choice
News Links


STReport's NEWS.STREPORT.COM Newsgroups

VOODOO 5 - 5500



by R. F. Mariano

This Video Card.... This 3D Video Card...... is absolutely stunning!

I had a Voodoo 3 3500 running in this machine and was as comfy as a bug in a rug with it.  It took long difficult hours to decide if I was going to pull it and install the Voodoo 5 5500.  In approximately 3 seconds my mind was made up.  After setting my Windows 2000 Professional machine back to vga video... I pulled the V3 card and installed the V5 card. Installation was a cakewalk.  I booted the machine and immediately, the card was found and I told the system where to find the drivers.. it found and happily installed them and asked me to re-boot.  All this, mind you, in 640x400 VGA resolution on a 21" Hitachi SuperScan 814 monitor.  Needless to say... the view was huge and ugly.

Upon re-booting... after the hash bar ran and a moment or two later, I heard the familiar rez-change click in the background and my desktop was back at 1280x1024 just as sharp and clear as could be.  I use my system for a myriad of things (including writing) so I must look for many areas to achieve near perfection in graphics and overall viewing. But I must add... the very first test is of course, with the eyes.... how it looks to me.  I booted an old favorite that I have many enjoyable hours with .. Quake 2.  Almost immediately, I saw such a change in the appearance of the game that I thought to myself... hmmm, those guys at ID really knew what they were doing.  God only knows what they were using for a video card when they wrote this game.  Folks, the difference in appearance was mind boggling.  The colors were more vivid, there were more colors, colors and texture where previously there was none.  The overall muddiness in the general rendition of the game was gone.  The game itself looked as if it were especially written for the voodoo5 card.

At this point I realized just how satisfied I was with the NEW Voodoo5.  I then booted Quake3 Arena and proceeded to re-configure the game. It now runs at 1280x1024. The smoothness and fine details rendering is superb.  Speed?  Did I hear you say Speed?  How's this... grab your shorts!  The game literally flies right along with NO hesitations at all.  Fast, Fast, Fast,  accentuated with SMOOTH,,,, In the past... with certain nVidia chips the puffs of smoke and explosions in Quake 3 Arena had a pixelated look to them ..they were grainy.  Now, it looks exactly like the real thing.  Shading and and shadows are just that. No longer are they in the classification of "close but no cigar".  With the Voodoo5 5500.. you get the entire humidor!

For the record; The machine in use here is:

  • Asus P3cd Motherboard

  • Two 600Mhz PIII CPU Chips

  • 256mb Memory - 2x 128mb RIMM Chips

  • Aureal SQ2500 Quad Surround Sound System running through a 150w (rms) Bose SS 4x with Subwoofer speaker system

  • Hitachi 21" SuperScan 814 monitor

  • Windows 2000 Professional OS

This system, assembled right here at STReport is, by everyday standards, a very fast system.  The components were chosen for specification and this method has proven to be the key to solid high performance.  If you want more info about the system, check into our recent Dream Machine Series... as an aside; in our opinion, after twelve years of testing and reviewing.. ASUS makes the best motherboards we have found so far.

3dfx... what can one say?  I've used their chips... then their cards all through the Voodoo series and I have found their cards to perform as well as any of the, twice the price, "so-called" Professional, Non-Gamer, Graphics Cards. 

The card was set up in a system running Win2k, Quake 3, Diablo2, Unreal Tournament and Quake 2.  That's on top of Photoshop 5.5 and a myriad of other programs in use on an everyday basis including Office 200 Premium featuring (for us at least) Front Page 2000... which is used extensively here.  This graphics card, the Voodoo 5 -5500 performed flawlessly across the boards.  From the smallest of print to most deadly confrontation in Arena, the 5500 did the job well above what was expected.  3dfx out did themselves with this card.

3dfx is, in our humble opinion, on top of the graphics scene and will continue to remain there. Whether its to purchase their products or invest in their future through their stock, you can't go wrong.  You will, in the long run be more than satisfied.

I felt it quite appropriate to add two files from 3fx to further the reader's knowledge of 3d technology.  This is needed if a well informed decision is to be made relative to the purchase of a true 3d card. 


VOODOO 5 5500

STReport Blue Ribbon Editor's Choice
Design - Support August 2000

General Features


64 MB of Graphics Memory

Enough memory to handle high-resolution gaming with large 32-bit textures and T-Buffer effects enabled.

667-733 Megapixels Per Second Fill Rate

Fill rate delivers frame rate. Virtually all games are limited by fill rate at resolutions of 1024x768 and above so the higher your fill rate the faster your game will perform. The Voodoo5 5500 AGP delivers the industry's highest fill rate, outside of the Voodoo5 6000, of course.

Real-time Full-Scene HW Anti-Aliasing

Removes annoying jaggies and flashing objects from the image to dramatically improve the visual quality of any title, new or old.

T-Buffer Digital Cinematic Effect: Motion blur

Can be used to smooth motion to improve image quality or to exaggerate motion for special effects.

T-Buffer Digital Cinematic Effect: Depth of Field Blur

Enhances reality by rendering a scene as if viewed through a real lens. Enables artistic depth of field effects often used in cinema, such as blurring all but the most important objects in a scene.

T-Buffer Digital Cinematic Effect: Soft Shadows

Adds soft edges to shadows to give them a much more realistic appearance.

T-Buffer Digital Cinematic Effect: Soft Reflections

Enables realistic reflections from semi-gloss surfaces like polished wood or stone and stainless steel.

FXT1™ and DirectX® Texture Compression

Dramatically reduces the size of texture images with no loss in visual quality; higher frame rates and better image quality. VSA-100 supports all five DXTC compression modes. Texture compression is critical for use of 2048x2048, 32-bit textures.

8-bit Palletized Textures

The most widely used form of texture compression; as with FXT1, palletized textures enable higher frame rates and enhanced image quality. Graphics cards that do not support palletized textures often experience severe game compatibility problems.

32-bit Rendering

Supports the highest possible image quality for the latest titles.

32-bit Textures

Allows use of the highest quality artwork seen in today's titles.

2k x 2k Textures

Allows you to enjoy the full impact of the most detailed artwork used by game developers; larger textures enable more complex, more stunning images.

24-bit Floating Point Depth Buffer (Z or W)

Virtually eliminates "Z aliasing," or having surfaces that should be visible being occluded and surfaces that should be occluded being visible.

8-bit Stencil Buffer

Improves image quality and realism by supporting a technique widely used by 3D content developers to create complex shapes such as shadows.

Fully integrated 128-bit 2D/3D/Video Accelerator

It's not just for gaming! Voodoo5 5500AGP enables outstanding 2D and video applications, too.


Allows users to run at insane resolutions and incredible refresh rate --2048x1536 at 85Hz for example.

The industry's most complete API support: DirectX, OpenGL, and Glide

Assures the industry's highest title compatibility.

DVD hardware assist: planar to packed-pixel conversion

Assists DVD titles in running at 30fps without dropped frames.

Windows 95, 98, ME and 2000

Allows you to run the Voodoo5 5500 AGP with standard operating systems.




White Paper on FXT1 Texture Compression Technology

3dfx Interactive


The development of advanced 3D content continues to push the envelope of on-screen visual realism, necessitating the use of more textures at higher resolutions to render more detailed and realistic images. Current titles are using two or more textures per object at resolutions up to 2048x2048 per texture and these requirements will continue to increase as 3D content developers strive for increased realism. Using more textures, however, also requires the support of an advanced and flexible technological foundation for handling them. 3dfx’s FXT1™ texture compression technology provides that foundation.

One of the issues when using large numbers of textures is simply that it takes much more memory to store those textures. Consider, for example, that a 2048x2048 32-bit per texel texture requires 16 Mbytes of texture storage space (and this is without even storing texture mipmaps!). Furthermore, this is only for a single high-resolution texture! It wasn’t long ago that 3D accelerators didn’t even have 16 Mbytes of total memory onboard. Texture compression is a powerful way to increase the amount of textures used without dramatically increasing the texture storage requirements.

Increasing the size and number of textures used in a scene also increases the amount of memory bandwidth required for texture lookup. Ultimately, as the amount of memory bandwidth required for texture accesses grows large, the overall fill-rate performance of the 3D accelerator is reduced and the resulting frame rate declines. One of the benefits of utilizing compressed textures is that the amount of memory bandwidth required for texturing is significantly reduced. As a result, games and applications taking advantage of texture compression can achieve higher sustained fill-rates and thus higher overall frame rates.

As a free, open-source, cross-platform compression algorithm, 3dfx’s FXT1™ texture compression technology allows content developers to create textures at higher resolutions and use more textures in a given scene while simultaneously reducing the memory bandwidth required for texturing. By reducing the amount of storage required for a given texture, more textures can be stored in a given amount of memory. As a result, more textures can be used in a rendered scene which can substantially improve overall visual quality. By reducing the memory bandwidth needed for texture transfers and processing, the overall fill-rate of a 3D accelerator is increased, which increases frame-rates and also generates a much more realistic, immersive 3D experience.

Advantages of FXT1™ Texture Compression:

-Increased total number of textures available for rendering

Texture compression technology reduces the amount of memory required to store a given texel, thus in turn reduces the amount of memory required to store an entire texture map. As a result of using texture compression, more textures can be stored in a given amount of texture memory. For example, a typical 3D accelerator storing a 256x256 texture in a 32 bit-per-texel format requires 256 Kbytes of texture memory storage. However, storing that same 256x256 texture in a compressed 4 bit-per-texel format requires only 32 Kbytes of texture memory storage, an effective savings of 224 Kbytes. Or, from a different perspective, the same amount of texture storage required to hold a single 256x256, 32 bit-per-texel texture can hold 8 unique 256x256, 4 bit-per-texel textures. These 8 unique textures can be used to dramatically improve the overall visual quality of a rendered scene.

Texture Size





4-bit FXT1™


6 KB

9 KB

13 KB

16 KB



18 KB

33 KB

49 KB

64 KB



66 KB

129 KB

193 KB

256 KB



258 KB

513 KB

769 KB

1024 KB



1,026 KB

2,049 KB

3,073 KB

4096 KB



4,098 KB

8,193 KB

12,289 KB

16384 KB


This table shows the memory requirements for various sizes and color depths of textures without compression. 
With FXT1™ texture compression, 32-bit images are reduced by a ratio of 8:1 without a perceivable loss in image quality.

-Higher resolution textures for better image quality

With limited on-board memory, first and second generation consumer 3D accelerators were limited to texture maps with a maximum resolution of 256x256 texels. These low-resolution textures look fine when viewed at a distance, but when viewed at a closer range the textures become blurred and lack fine detail. However, utilizing higher resolution textures to improve visual quality can cause texture memory requirements to be too large to be economical. By utilizing texture compression, higher resolution textures can now be utilized to offer significantly improved visual realism in economical amounts of texture memory. Consider our example 256x256, 32 bit-per-texel texture map. The same amount of memory required to store this texture map, 256 Kbytes, can hold a single 4 bit-per-texel texture map at a resolution of 1024x512. Utilizing larger texture maps can significantly enhance the overall visual quality of a rendered scene.

-More textures per polygon for advanced effects

With support for multiple textures per surface in all of the major 3D APIs, content developers are now using two or more textures per surface to increase image quality and create special effects. By using more than one texture per surface, effects like advanced real-time lighting, specular highlights, and bump mapping are possible, without increasing the total triangle count for the scene or overwhelming the available memory bandwidth.

With multiple textures, advanced lighting effects like spotlights and shadows can be created very easily and without using geometrically complex models. Lighting effects using multi-texturing hardware are typically performed by blending a light map (a black and white image that emulates the differences in light intensity) with the base texture of an object. When these two textures are blended, the user sees all the lighting detail without the performance burden caused by alternative techniques which utilize significantly higher geometric models and complex lighting equations.

Texture compression benefits these multi-texturing effects by decreasing the total bandwidth needed to transfer the extra texture data used for the second texture . Additionally, the resultant memory savings which comes from using compressed textures allows the application to use multi-texturing techniques on many more objects in the scene. Consider that a typical 512x512 texture at 32 bits-per-texel would require 1MB of texture storage space. If the 3D scene contained just eight objects with two textures each, current generation cards (with only 16MB of memory) would use up all available bandwidth and storage space with no room left over for mipmaps. With 4-bit-per-texel texture compression, a total of 40 or more objects could be multi-textured, including mip maps, before memory storage is depleted. This allows multi-texturing for entire rooms and all of the objects or characters in a given scene.


-Lower bandwidth requirements for better fill-rate performance

Fill-rate is the number of pixels or texels that can be drawn in a given period of time. The higher the fill-rate of a 3D accelerator, the higher frame rates it will produce for a given piece of content. Of course higher frame rates are desirable as they create smooth 3D rendering without the jerkiness associated with lower frame rates. Overall, fill-rate performance is directly affected by the amount of texture memory bandwidth available. In the case of a 32-bit texture, 8 times the amount of texture memory bandwidth is required to render a given texel when compared to a 4-bit texture. This dramatic reduction of required texture memory bandwidth, as a result of utilizing texture compression, results in much higher fill-rates and frame rates, thus creating a much more enjoyable, immersive 3D experience.

FXT1™ texture compression works by dividing an image into multiple 4x4 and/or 4x8 texel blocks. Each texture (including opaque and transparent textures) is compressed to an average of four bits-per-texel. In the FXT1™ texture compression scheme, four different compression algorithms are used, with the best one chosen per block to generate the highest quality result. The four algorithms used are:

CC_MIXED (similar to other texture compression schemes): A 4x4 texel block is represented by two bits-per-texel for opaque textures. Additionally, each block has two 16-bit colors stored in an RGB 565 format. The two RGB 565 colors and two additional colors (created by interpolating between the two RGB 565 colors) form the primary colors for this texel block and its associated four color lookup table. A 2-bit index is used to determine which color from the lookup table will be used for each texel in the 4x4 block. Transparent textures are created by making one of the four colors transparent.

CC_HI (best for spatial resolution): A 4x8 texel block is represented by three bits-per-texel for opaque and transparent textures. Each block stores two 15-bit colors in an RGB 555 format. The two RGB 555 colors and five additional colors (created by interpolating between the two RGB 555 colors) form the primary colors for this texel block. Additionally, an eighth color is defined to be the transparent color. A 3-bit index is used to determine which color from the 8-entry lookup table will be used for each texel in the 4x8 block.

CC_CHROMA (good at complex color areas): A 4x8 texel block is represented by two bits-per-texel for opaque textures. Each block stores four 15-bit colors in an RGB 555 format. All four colors are used directly with no interpolation to form a 4-entry lookup table. The 2-bit index assigned to each texel in the block is used to determine which of the four colors is assigned to each individual texel. Note that Colors4 only applies to opaque textures, as it does not support transparency.

CC_ALPHA (gives the best control over complex alpha transparencies at four bits-per-texel): A 4x8 texel block is represented by two bits-per-texel for opaque and transparent textures. Each block stores three 20-bit colors stored in a 5555 format. The first and second 20-bit colors are used for the primary colors of the left 4x4 block, while the second and third colors are used for the primary colors of the right 4x4 block. Two additional colors are created in each block by interpolating between the two primary colors for that block. A 2-bit index is assigned to each texel in the block and a lookup table is used to determine which color is applied to each texel.

-Decompression (also known as Decoding):

FXT1™ texture decompression happens during the texture mapping process and is implemented in the 3D hardware accelerator. A 2-bit field is stored in each block and is used to determine which of the 4 compression schemes was utilized by the compression algorithm for best visual quality. Depending on which algorithm is used for a given block, the proper decompression logic is applied to generate decoded 32-bit texels which can then be used by the texture mapping hardware.

Advantages of FXT1™ texture compression as compared to S3TC™

-Free, open source tool set:

3dfx wants to encourage software content developers and 3D hardware accelerator companies (also known as "IHVs") to innovate with new encoding and decoding schemes that offer even greater compression ratios, better image quality, or ideally both. To that end, the compression and decompression tools, and source code are freely available. This will allow developers to use the FXT1™ compression technology across multiple platforms. Furthermore, the free license will provide IHV’s with the opportunity to implement FXT1™ texture compression in future hardware designs, completely free of any license or royalty fees.

-Multiple encoding schemes per image for better quality

FXT1™ creates images with higher quality than other texture compression schemes by using four compression techniques for each texture (compared to only a single technique in other compression schemes). This allows the encoder to be more accurate in reproducing specific portions of an image and/or different types of images as the best possible technique is applied to each texel block.

-Better Compression Ratio for textures with multi-bit Alpha

Unlike S3TC™ which uses an 8-bit compression format when compressing textures with multi-bit alpha components (alpha is used for transparency information), FXT1™ uses a 4-bit format for the greatest possible compression ratio. As a result, the compression ratio of the FXT1™ compression algorithm is twice that of the S3TC algorithm when compressing 16 or 32-bit textures which include alpha information. This substantially increases the number of textures which can be stored in a given amount of memory, and also reduces the amount of bandwidth required for texturing.

-Cross Platform API support

All major 3D APIs support FXT1™ texture compression. Textures can be pre-compressed and stored on CD or they can be compressed in real-time by using the provided tools and source code for FXT1™ encoding. In Direct3D™, when an application creates a textured surface in memory, the application tells the hardware if there are any special codes used to designate a compressed image surface, called FOURCC codes, associated with that surface. If so, the hardware reads this code and knows that the surface must be decoded prior to being moved to the frame buffer. In OpenGL (on all platforms, including Windows, Linux and Macintosh), texture surfaces are marked with an FXT1™code in the file header. When a surface is created in memory, the surface has a flag which designates it as a compressed texture. The OpenGL driver recognizes this flag when the proper extension is implemented in the OpenGL driver. The texture is then decoded by the hardware, prior to being used internally. Glide provides native support for FXT1™ based textures; automatically recognizing FXT1™ encoded textures and decoding them accordingly in hardware.

Using FXT1™ Texture Compression

3dfx provides free tools and associated source code for encoding and decoding textures in the FXT1™ format. One tool is a command line utility which allows developers to convert back and forth between the FXT1™ format and other standard image formats, performing both FXT1™ compression and decompression. A second tool is an Adobe PhotoShop™ plug-in, which allows artists to work on a texture image within PhotoShop and then both export and import images in the FXT1™ image format. For developers wishing to do real-time compression in the FXT1™ format, free source code is available for writing an encoder as part of an application.


FXT1™ texture compression is provided free to software developers and 3D hardware accelerator companies who want to create and accelerate content utilizing many more texture images and higher resolution textures, all at the highest possible image quality. FXT1™ texture compression provides equal to or better compression ratios than any available hardware compression scheme. With up to four compression schemes used in the encoding of each texture, FXT1™ provides the most accurate texture compression available, providing little or no loss in image quality. Decoding FXT1™ textures can be done transparently in the 3D hardware at run-time, or decoded in software and converted to another hardware-supported texture format. In addition, cross platform 3D API (Glide, OpenGL, and Direct3D) support for FXT1™ allows developers to use it on Windows, Macintosh and Linux platforms without any additional coding by the developer. Finally, 3dfx provides free tools, and the associated source code, to encode and decode compressed textures to allow the user community to create newer and better encoding algorithms. These tools are currently available to developers registered with the 3dfx Developer Program. For additional information, please contact

on to page two 


Email Us Questions? Comments?
 Email Us!
Legal Information and Privacy Policy © 1985-2014 STReport Magazine