Did you know that Rage Pro supports texture compression and CIF games use VQ compression?
Maybe thats one reaseon RagePro / RageXL and Tomb Raider gets good framerate@800x600?
https://www.rage3d.com/board/index.php?thread … mpression.2197/
ATI Rage VQ compression refers to vector quantization, a lossy compression method used by the ATI Rage Pro and other early ATI 3D graphics cards to reduce texture sizes for improved performance. This technology allowed for the storage of textures in a more compressed format, with the graphics card then decompressing and rendering the textures in real-time.
Function: VQ was a lossy compression technique used for 3D textures on the Rage Pro chip.
Benefit: By compressing texture data, it allowed for more textures to be stored in memory or loaded faster, which could significantly improve 3D graphics performance at the time.
How it worked: The compression worked by grouping similar colors into codebooks. Instead of storing the full texture pixel data, the GPU would store a reference to the codebook, making the file size smaller.
More VQTC info and also GTA3andGTA4@Rage3
Doubts about ATI rage, SiS and others with 8MB
ATI's use of VQ compression likely refers to its 3Dc compression algorithm, which was a dedicated implementation of vector quantization for textures in its graphics hardware. This technique was used to compress texture data to reduce bandwidth requirements and improve performance by dividing textures into blocks and representing them with fewer bits.
How it works
Vector Quantization (VQ): A classical signal processing technique that compresses data by dividing a large set of vectors into groups. Each group is represented by a single "centroid" or representative vector.
ATI's implementation (3Dc): ATI developed this algorithm for its graphics hardware to efficiently compress texture data.
Reduced bandwidth: By using VQ, 3Dc significantly reduced the amount of data that needed to be transferred from memory to the graphics processor.
Performance improvement: The compression allowed for faster loading of textures, which is crucial for real-time rendering in video games and other graphics applications.
Lossy compression: Like most VQ methods, meaning some information is lost during the compression process. The goal is to minimize the visual impact of this data loss.
Hardware implementation: By implementing the algorithm directly in its graphics hardware, ATI could decompress the textures quickly without taxing the CPU.
Rage Pro (LT Pro) VQ 8:1
http://www.bitsavers.org/components/ati/RAGE_ … v_1.01_1997.pdf
"Rage Pro VQ" most likely refers to the ATI 3D Rage Pro graphics processor, which supported Vector Quantization (VQ) for texture compression. The 3D Rage Pro was a graphics chip released in 1997 that was used in various graphics cards for its capabilities in 3D graphics.
ATI 3D Rage Pro
Architecture: Based on the Rage 3 architecture.
Release: March 1, 1997.
Production Process: 350 nm.
Key Features:Supported DirectX 6.0.
Included hardware support for texture compression using Vector Quantization (VQ), which allowed for an 8:1 compression ratio to save memory bandwidth and capacity.
Vector Quantization (VQ)
Function: A texture compression algorithm used by the Rage Pro chip.
Benefit: Greatly reduced memory bandwidth and capacity requirements for texturing, allowing for more detailed textures in games.
Limitation: Required a proprietary API (CIF) to use.
Here is one proof that Rage XL has VQ, but here its 4:1, not 8:1 like Rage Pro and LT Pro has?
https://theretroweb.com/chip/documentation/at … 3b508495717.pdf
Whats the difference between VQ 4:1 and VQ 8:1 compression?
"Rage Pro VQ 8:1" refers to an 8:1 texture compression method using Vector Quantization (VQ) as a feature of the ATI 3D RAGE PRO graphics processor. This compression reduced memory requirements and bandwidth, allowing developers to use larger, richer textures in games and applications during the late 1990s.
Texture Compression: VQ compression is a lossy method that significantly reduces the memory and bandwidth needed for textures by encoding blocks of pixels into a smaller set of "codebook" vectors.
8:1 Compression: The "8:1" ratio means that the compressed texture occupies only one-eighth the space of the original, uncompressed texture.
Impact: This technology helped developers create more visually detailed content on hardware with limited memory, though it was only usable through the proprietary ATI CIF API.
"ATi Rage VQ 4:1" refers to the vector quantization (VQ) texture compression technology used in the ATi RAGE series of graphics accelerators, which allowed for a compression ratio of up to 4:1. This technology was a feature on cards like the ATi RAGE XL, helping them to improve 3D performance by reducing the amount of texture data that needed to be processed and stored, a significant improvement for its time.
VQ 4:1 Texture Compression: This was a method of compressing texture data to a quarter of its original size (a 4:1 ratio) by using a "vector quantization" technique.
Performance Improvement: By compressing textures, the graphics accelerator could use less memory bandwidth and store more textures in its cache, leading to better performance in 3D applications.
Historical Context: This was a key feature in the era of the ATi RAGE graphics accelerators, which were designed for 3D and 2D acceleration. The technology helped the cards perform better than they would have without this compression method.
4:1 and 8:1 are the same, its only how you look at it, with or without palletized textures?
https://dri.freedesktop.org/wiki/TextureCompression/
ATI used Vector Quantization (VQ) for texture compression on its early graphics cards, such as the 3D RAGE PRO
and developed a specialized version called 3Dc for its later hardware.
ATi 3Dc and X800
https://www.tomshardware.com/reviews/ati,802-7.html
VQ or VQC or VQTC should work out of ATi 3DCIF API, Rage128 supports VQ and it does not have 3DCIF API support.
