GeForce 4 series
Fourth generation of Nvidia GeForce graphics processors
Nvidia released the GeForce 4 series in early 2002 as its fourth generation of graphics processing units. The lineup split into high-performance Ti models and budget MX chips, introducing improved memory controllers and video playback features for desktops and laptops.
- Manufactured on TSMC 150 nm process
- Introduced Direct3D 8.1 support in Ti models
- MX line featured Video Processing Engine
- Ti4200 offered strong price-to-performance ratio
- Supported AGP-8X in late 2002 revisions
What it is
The GeForce 4 series represents the fourth generation of graphics processing units from Nvidia. Announced in February 2002, this family divides into two distinct groups. The Ti line targets high performance with advanced shader support, while the MX line serves budget systems with improved video capabilities. Both families use the Kelvin microarchitecture and were manufactured using a 150 nanometre process by TSMC. The series introduced features like hardware anti-aliasing and better multi-monitor support, bridging the gap between older DirectX 7 hardware and emerging DirectX 9 standards.
How it came about
Nvidia launched the initial Ti4400 and Ti4600 models in early 2002 to compete with ATI Radeon cards. The company delayed the cheaper Ti4200 to clear existing GeForce 3 inventory, a move that allowed competitors to gain market share. Late in the year, Nvidia updated the cores to support AGP-8X, releasing the Ti4800 and MX440-8X. The mobile segment saw the introduction of GeForce4 Go chips derived from the MX line, followed by a Ti-based mobile variant that suffered from high thermal output. Driver support continued for several years, with final releases for older Windows systems arriving in 2006.
Inside the chip
The Ti models feature the nFinite FX Engine II, enabling Vertex Shader 1.1 and Pixel Shader 1.3 compliance under Direct3D 8.1. They utilize the Lightspeed Memory Architecture II to improve bandwidth efficiency. The MX chips lack programmable shaders but include the Video Processing Engine for hardware MPEG-2 decoding. Both lines support TwinView for dual monitors and Accuview for anti-aliasing. The MX architecture draws heritage from the GeForce 2 but adds a 128-bit DDR memory controller to resolve earlier bandwidth bottlenecks. Later revisions added AGP-8X interface support via the NV18 and NV28 cores.
How it was made
TSMC manufactured all chips in the series using a 150 nanometre fabrication process. The product codes include NV17 for the base MX, NV25 for the initial Ti, and NV28 for the AGP-8X Ti update. Mobile variants used NV17M and NV28M designations. A later PCI Express model, the PCX 4300, used an NV19 core with a bridge chip to connect the native AGP interface to the newer bus standard. Production volumes focused heavily on the OEM market for the MX line.
Where it went
Desktop PCs utilized these cards for gaming and general graphics acceleration. The Ti4200 became a popular choice for value-oriented gamers due to its longevity and price point. OEMs widely adopted the MX440 for mass-market computers, where it outperformed integrated graphics solutions. The mobile Go variants powered laptops, though the Ti-based mobile chip faced adoption issues due to heat and power consumption. Professional applications like AutoCAD also benefited from the MX line’s optimized performance in non-shader tasks.
Versions and family
The Ti family includes the 4200, 4400, 4600, 4800 SE, and 4800. The MX family comprises the 420, 440, and 460, along with later 8X and SE revisions. Mobile versions include the 420 Go, 440 Go, and 460 Go from the MX line, plus the 4200 Go from the Ti line. The PCX 4300 serves as a PCI Express variant of the MX440SE. Each model differentiates primarily through core clock speeds, memory bus width, and frame buffer size.
Why it matters
The Ti4200 established a strong reputation for price-to-performance balance, remaining competitive against early DirectX 9 cards. The MX series secured a dominant position in the OEM sector, outselling many discrete competitors despite criticism for its naming convention. The series helped transition the market from fixed-function pipelines to programmable shaders, although the MX line retained legacy architecture. Its video decoding engines set a new baseline for multimedia playback on personal computers.
Problems and reception
Critics viewed the MX naming as misleading because it lacked the shader hardware of the Ti line and predecessor GeForce 3. Gamers initially avoided the MX440, but its widespread OEM presence forced developers like id Software to support it. The Ti4400 struggled to find a market niche between the high-end Ti4600 and the value-focused Ti4200. Mobile users reported poor battery life with the Ti-derived Go chip due to missing power-saving circuitry.




Questions people ask about GeForce 4 series
What is the difference between GeForce 4 Ti and MX?
The Ti series supports Direct3D 8.1 with programmable vertex and pixel shaders, targeting high performance. The MX series lacks these shaders but includes a Video Processing Engine for better movie playback and targets budget systems.
Which manufacturing process was used for these chips?
All models in the GeForce 4 series, including desktop and mobile variants, were manufactured by TSMC using a 150 nanometre process node.
Did the GeForce 4 support PCI Express?
Most models used AGP interfaces. The PCX 4300 was a later variant that supported PCI Express by using a bridge chip to connect the native AGP core to the newer bus standard.
Why was the GeForce 4 MX criticized?
Enthusiasts felt the name was misleading because the MX lacked the advanced shader hardware found in the GeForce 3 and GeForce 4 Ti, despite being part of the same generation branding.
Good reading and sources
- GeForce RTX 20 Series Graphics Cards and LaptopsNvidia, Article
- Checking your browserAnandTech, Article
- PC Graphics Beyond XBOXTom's Hardware, Article
- GeForce 4 series on WikipediaEncyclopedia
- More photos on Wikimedia CommonsPhotos
- Wikidata recordOpen data
The facts on this page come from the Wikipedia article "GeForce 4 series", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.
Like GeForce 4 series
All graphics processors
WorldGeForce 3 series
Third-generation Nvidia GPU with programmable shadersGPU · GPU architecture · 2001
More from Nvidia
All of them
🇯🇵 JapanSPARC64 V
Single-core SPARC V9 server microprocessor by FujitsuCPU · Other · 2001