Hubbry Logo
GeForce 4 seriesGeForce 4 seriesMain
Open search
GeForce 4 series
Community hub
GeForce 4 series
logo
7 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
GeForce 4 series
GeForce 4 series
GeForce 4 series
View on Wikipedia
from Wikipedia

Nvidia GeForce 4 series
Logo
Release dateFebruary 6, 2002; 23 years ago (February 6, 2002)
CodenameNV17, NV18, NV19, NV25, NV28
ArchitectureKelvin
Cards
Entry-levelMX
Mid-rangeTi 4200, Ti 4400, Ti 4800 SE
High-endTi 4600, Ti 4800
API support
Direct3DDirect3D 7.0 NV1x
Direct3D 8.0a NV2x
Vertex Shader 1.1
Pixel Shader 1.3
OpenGLOpenGL 1.3
History
PredecessorGeForce 3 series
SuccessorGeForce FX series
Support status
Unsupported

The GeForce 4 series (codenames below) refers to the fourth generation of Nvidia's GeForce line of graphics processing units (GPUs). There are two different GeForce4 families, the high-performance Ti family (NV25), and the budget MX family (NV17). The MX family spawned a mostly identical GeForce4 Go (NV17M) family for the laptop market. All three families were announced in early 2002; members within each family were differentiated by core and memory clock speeds. In late 2002, there was an attempt to form a fourth family, also for the laptop market, the only member of it being the GeForce4 4200 Go (NV28M) which was derived from the Ti line.

GeForce4 Ti

[edit]
GeForce4 Ti 4800 (NV28 Ultra) GPU
Albatron GeForce4 Ti 4800SE
GeForce4 Ti 4600 Mac

Architecture

[edit]

The GeForce4 Ti (NV25) was launched in February 2002[1] and was a revision of the GeForce 3 (NV20). It was very similar to its predecessor; the main differences were higher core and memory clock rates, a revised memory controller (known as Lightspeed Memory Architecture II/LMA II), Direct3D 8.1 support with up to Pixel Shader 1.3,[2][3][4] an additional vertex shader (the vertex and pixel shaders were now known as nFinite FX Engine II), hardware anti-aliasing (Accuview AA), and DVD playback.[1] Legacy Direct3D 7-class fixed-function T&L was now implemented as vertex shaders.[3] Proper dual-monitor support (TwinView) was also brought over from the GeForce 2 MX.[5] The GeForce 4 Ti was superior to the GeForce 4 MX in virtually every aspect save for higher production cost, although the MX had the Nvidia VPE (video processing engine) which the Ti lacked.

Lineup

[edit]

The initial two models were the Ti4400 (US$299) and the top-of-the-range Ti4600 (US$399). At the time of their introduction, Nvidia's main products were the entry-level GeForce 2 MX, the midrange GeForce4 MX models (released the same time as the Ti4400 and Ti4600), and the older but still high-performance GeForce 3 (demoted to the upper mid-range or performance niche).[1] However, ATI's Radeon 8500LE (9100) (a slower-clocked version of the flagship Radeon 8500) was somewhat cheaper than the Ti4400, and outperformed its price competitors, the GeForce 3 Ti200 and GeForce4 MX 460. The GeForce 3 Ti500 (US$299 prior to the GeForce4 release) was rendered obsolete quickly as it could not be produced cheap enough to justify a further price drop, though it filled the performance gap between the GeForce 3 Ti200 and the GeForce4 Ti4400.[6]

In consequence, Nvidia rolled out a slightly cheaper model: the Ti4200. Although the 4200 was initially supposed to be part of the launch of the GeForce4 line, Nvidia had delayed its release to sell off the soon-to-be discontinued GeForce 3 Ti500 chips.[7] In an attempt to prevent the Ti4200 damaging the Ti4400's sales, Nvidia set the Ti4200's memory speed at 222 MHz on the models with a 128 MiB frame buffer—a full 53 MHz slower than the Ti4400 (all of which had 128 MiB frame buffers). Models with a 64 MiB frame buffer were set to 250 MHz memory speed. This tactic didn't work however, for two reasons. Firstly, the Ti4400 was perceived as being not good enough for those who wanted top performance (who preferred the Ti4600), nor those who wanted good value for money (who typically chose the Ti4200), causing the Ti4400 to fade into obscurity.[8] Furthermore, some graphics card makers simply ignored Nvidia's guidelines for the Ti4200, and set the memory speed at 250 MHz on the 128 MiB models anyway.[9]

Then in late 2002, the NV25 core was replaced by the NV28 core, which differed only by addition of AGP-8X support. The Ti4200 with AGP-8X support was based on this chip, and sold as the Ti4200-8X. A Ti4800SE replaced the Ti4400 and a Ti4800 replaced the Ti4600 respectively when the 8X AGP NV28 core was introduced on these.[10][11]

The only mobile derivative of the Ti series was the GeForce4 4200 Go (NV28M), launched in late 2002.[12] The solution featured the same feature-set and similar performance compared to the NV28-based Ti4200, although the mobile variant was clocked lower. It outperformed the Mobility Radeon 9000 by a large margin, as well as being Nvidia's first DirectX 8 laptop graphics solution. However, because the GPU was not designed for the mobile space, it had thermal output similar to the desktop part. The 4200 Go also lacked power-saving circuitry like the MX-based GeForce4 4x0 Go series or the RV250-based Mobility Radeon 9000. This caused problems for notebook manufacturers, especially with regards to battery life.[13]

Performance

[edit]

The GeForce4 Ti outperformed the older GeForce 3 by a significant margin.[1] The competing ATI Radeon 8500 was generally faster than the GeForce 3 line, but was overshadowed by the GeForce 4 Ti in every area other than price and more advanced pixel shader (1.4) support.[1] Nvidia, however, missed a chance to dominate the upper-range/performance segment by delaying the release of the Ti4200 and by not rolling out 128 MiB models quickly enough; otherwise the Ti4200 was cheaper and faster than the previous top-line GeForce 3 Ti500 and Radeon 8500.[14] Besides the late introduction of the Ti4200, the limited release 128 MiB models of the GeForce 3 Ti200 proved unimpressive, letting the Radeon 8500LE and even the full 8500 dominated the upper-range performance for a while.[15] The Matrox Parhelia, despite having several DirectX 9.0 capabilities and other innovative features, was at most competitive with the GeForce 3 and GeForce 4 Ti 4200, but it was priced the same as the Ti 4600 at US$399.

The GeForce 4 Ti4200 enjoyed considerable longevity compared to its higher-clocked peers. Debuting at half the cost of the 4600 (US$199 versus US$399), the 4200 remained the best balance between price and performance until the launch of the short-lived DirectX 9 ATI Radeon 9500 Pro at the end of 2002.[16] The Ti4200 still managed to hold its own against several next generation DirectX 9 compliant GPUs released in late 2003, outperforming the GeForce FX 5200 and the midrange FX 5600, and performing similarly to the mid-range Radeon 9600 Pro (ATI's permanent successor to the Radeon 9500 Pro) in some situations.[17][18][19]

GeForce4 MX

[edit]

Architecture

[edit]
GeForce4 MX420

Though its lineage was of the past-generation GeForce2 (NV11 and NV15), the GeForce4 MX (NV17) did incorporate bandwidth and fill rate-saving techniques, dual-monitor support, and a multi-sampling anti-aliasing unit from the Ti series (NV25); the improved 128-bit DDR memory controller was crucial to solving the bandwidth limitations that plagued the GeForce 256 (NV10) and GeForce2 lines. It also owed some of its design heritage to Nvidia's high-end CAD products, and in performance-critical non-game applications it was remarkably effective. The most notable example is AutoCAD, in which the GeForce4 MX returned results within a single-digit percentage of GeForce4 Ti cards several times the price.

Many criticized the GeForce4 MX name as a misleading marketing ploy since it was less advanced than the preceding GeForce3 (NV20). In the features comparison chart between the Ti and MX lines, it showed that the only "feature" that was missing on the MX was the nfiniteFX II engine—the DirectX 8 programmable vertex and pixel shaders.[20] However, the GeForce4 MX was not a GeForce4 Ti with the shader hardware removed, as the MX's performance in games that did not use shaders was considerably behind the Ti.

In motion-video applications, the GeForce4 MX offered new functionality. It (and not the GeForce4 Ti) was the first GeForce member to feature the Nvidia VPE (video processing engine). It was also the first GeForce to offer hardware-iDCT and VLC (variable length code) decoding, making VPE a major upgrade from Nvidia's previous HDVP. In the application of MPEG-2 playback, VPE could finally compete head-to-head with ATI's video engine.

Lineup

[edit]
GeForce4 MX440

There were 3 initial models: the MX420, the MX440 and the MX460. The MX420 had only Single Data Rate (SDR) memory and was designed for very low end PCs, replacing the GeForce2 MX100 and MX200. The GeForce4 MX440 was a mass-market OEM solution, replacing the GeForce2 MX/MX400 and GeForce2 Ti. The GeForce4 MX460 was initially meant to slot in between the MX440 and the Ti4400, while the release of the Ti4200 was held back.

In terms of 3D performance, the MX420 performed only slightly better than the GeForce2 MX400 and below the GeForce2 GTS. However, this was never really much of a problem, considering its target audience. The nearest thing to a direct competitor the MX420 had was ATI's Radeon 7000. In practice its main competitors were chipset-integrated graphics solutions, such as Intel's 845G and Nvidia's own nForce 2, but its main advantage over those was multiple-monitor support; Intel's solutions did not have this at all, and the nForce 2's multi-monitor support was much inferior to what the MX series offered.

NVIDIA GeForce PCX 4300

The MX440 performed reasonably well for its intended audience, outperforming its closest competitor, the ATI Radeon 7500, as well as the discontinued GeForce2 Ti. Despite harsh criticism by gaming enthusiasts, the MX440 was successful in the PC OEM market as a replacement for the GeForce2 MX. Priced about 30% above the GeForce2 MX, it provided better performance, the ability to play a number of popular games that the GeForce2 could not run well—above all else—to the average non-specialist it sounded as if it were a "real" GeForce4—i.e., a GeForce4 Ti. While John Carmack initially warned gamers not to purchase the GeForce 4 MX440, its somewhat widespread adoption compelled id Software to make it the only DirectX 7.0 GPU supported by Doom 3.[21][22] When ATI launched its Radeon 9000 Pro in September 2002, it performed about the same as the MX440, but had crucial advantages with better single-texturing performance and proper support of DirectX 8 shaders. However, the 9000 was unable to break the MX440's entrenched hold on the OEM market. Nvidia's eventual answer to the Radeon 9000 was the GeForce FX 5200, but despite the 5200's DirectX 9 features it did not have a significant performance increase compared to the MX440 even in DirectX 7.0 games. This kept the MX440 in production while the 5200 was discontinued.

One of the fastest DirectX 7.0 compliant GPUs, the MX460's performance was similar to the discontinued GeForce 2 Ultra and the existing GeForce3 Ti200 (the remaining available member of the GeForce 3 family). However, ATI released the Radeon 8500LE which outperformed its price competitors, the GeForce 3 Ti200 and GeForce4 MX 460. ATI's move in turn compelled Nvidia to roll out the Ti4200 earlier than planned, also at a similar price to the MX 460, and soon afterwards discontinuing the Ti200. The Ti200, 8500LE, and Ti4200 were all DirectX 8.0 compliant while having similar market pricing to the MX460, while the 8500LE and Ti4200 also provided significantly better performance than the MX460, prevented the MX460 from ever being popular compared to the other GeForce 4 MX releases.

The GeForce4 Go was derived from the MX line and it was announced along with the rest of the GeForce4 Ti and MX lineup in early 2002. There was the 420 Go, 440 Go, and 460 Go. However, ATI had beaten them to the market with the similarly performing Mobility Radeon 7500, and later the DirectX 8.0 compliant Mobility Radeon 9000. (Despite its name, the short-lived 4200 Go (NV28M) is not part of this lineup, it was instead derived from the Ti line.)

Like the Ti series, the MX was also updated in late 2002 to support AGP-8X with the NV18 core. The two new models were the MX440-8X, which was clocked slightly faster than the original MX440, and the MX440SE, which had a narrower memory bus, and was intended as a replacement of sorts for the MX420. The MX460, which had been discontinued by this point, was never replaced. Another variant followed in late 2003—the MX 4000, which was a GeForce4 MX440SE with a slightly higher memory clock.

The GeForce4 MX line received a third and final update in 2004, with the PCX 4300, which was functionally equivalent to the MX 4000 but with support for PCI Express. In spite of its new codename (NV19), the PCX 4300 was in fact simply an NV18 core with a BR02 chip bridging the NV18's native AGP interface with the PCI-Express bus.

GeForce4 model information

[edit]
Main article: Comparison of Nvidia graphics processing units
Further information: Kelvin (microarchitecture)
  • All models are manufactured via TSMC 150 nm manufacturing process
  • All models support nView
Model Launch
Code name
Transistors (million)
Die size (mm2)
Core clock (MHz)
Memory clock (MHz)
Core config[a]
 Fillrate Memory Supported API version
Performance (GFLOPS
FP32)
TDP (Watts)
MOperations/s
MPixels/s
MTexels/s
MVertices/s
Size (MB)
Bandwidth (GB/s)
Bus type
Bus width (bit)
GeForce4 MX IGP + nForce2 October 1, 2002 NV1F ? ? FSB 250 133
200 		2:0:4:2 500 500 1,000 125 Up to 128 system RAM 2.128
6.4 		DDR 64
128 		7.0 1.2 1.000 ?
GeForce4 MX420 February 6, 2002 NV17 29[23] 65 AGP 4x
PCI 		166 64 2.656 SDR
DDR 		128 (SDR)
64 (DDR) 		14
GeForce4 MX440 SE 2002 133
166[24] 		500[24]

1000

64
128 		2.128 5.312[24] DDR 64 128[24] 13
GeForce MX4000 December 14, 2003 NV18B 29 65 AGP 8x
PCI 		166 1000 2.656 64 14
GeForce PCX4300 February 19, 2004 PCIe x16 128 16
GeForce4 MX440 February 6, 2002 NV17 29 65 AGP 4x
PCI 		275 200 550 550 1,100 137.5 64
128 		6.4 128 1.100 18
GeForce4 MX440 8x September 25, 2002 NV18 29[25] 65 AGP 8x
PCI 		166
250 		2.656[26]
8.0 		64
128 		19
GeForce4 MX460 February 6, 2002 NV17 29 65 AGP 4x
PCI 		300 275 600 600 1,200 150 8.8 128 1.200 22
GeForce4 Ti4200 April 16, 2002 NV25 63[27] 142 AGP 4x 250 222 (128 MB)
250 (64 MB) 		4:2:8:4 1,000 1,000 2,000 125 7.104 (128 MB)
8.0 (64 MB) 		8.1 1.3 15.00 33
GeForce4 Ti4200 8x September 25, 2002 NV28 63[28] 101[29] AGP 8x 250 8.0 34
GeForce4 Ti4400 February 6, 2002 NV25 63 142 AGP 4x 275 275 1,100 1,100 2,200 137.5 128 8.8 16.50 37
GeForce4 Ti4400 8x
(Ti4800SE[b]) 		January 20, 2003 NV28 63 101 AGP 8x 38
GeForce4 Ti4600 February 6, 2002 NV25 63 142 AGP 4x 300 325 1,200 1,200 2,400 150 10.4 18.00 ?
GeForce4 Ti4600 8x
(Ti4800[c]) 		January 20, 2003 NV28 63 101 AGP 8x 43
  1. ^ Pixel shaders: vertex shaders: texture mapping units: render output units
  2. ^ GeForce4 Ti4400 8x: Card manufacturers utilizing this chip labeled the card as a Ti4800SE. The surface of the chip has "Ti-8x" printed on it.
  3. ^ GeForce4 Ti4600 8x: Card manufacturers utilizing this chip labeled the card as a Ti4600, and in some cases as a Ti4800. The surface of the chip has "Ti-8x" printed on it, as well as "4800" printed at the bottom.
Model Features
nFiniteFX II Engine Video Processing Engine (VPE)
GeForce4 MX420 No Yes
GeForce4 MX440 SE No Yes
GeForce4 MX4000 No Yes
GeForce4 PCX4300 No Yes
GeForce4 MX440 No Yes
GeForce4 MX440 8X No Yes
GeForce4 MX460 No Yes
GeForce4 Ti4200 Yes No
GeForce4 Ti4200 8x Yes No
GeForce4 Ti4400 Yes No
GeForce4 Ti4400 8x Yes No
GeForce4 Ti4600 Yes No
GeForce4 Ti4600 8x Yes No

GeForce4 Go mobile GPU series

[edit]
  • Mobile GPUs are either soldered to the mainboard or to some Mobile PCI Express Module (MXM).
  • All models are made via 150 nm fabrication process
Model Launch
Core clock (MHz)
Memory clock (MHz)
Core config[a]
 Fillrate Memory API support
MOperations/s
MPixels/s
MTexels/s
MVertices/s
Size (MB)
Bandwidth (GB/s)
Bus type
Bus width (bit)
GeForce4 Go 410 February 6, 2002 NV17M AGP 8x 200 200 2:0:4:2 400 400 800 0 16 1.6 SDR 64 7.0 1.2
GeForce4 Go 420 400 32 3.2 DDR
GeForce4 Go 440 220 440 440 440 880 64 7.04 128
GeForce4 Go 460 October 14, 2002 250 500 500 500 1000 8
GeForce4 Go 488 NV18M 300 550 600 600 1200 8.8
GeForce4 Go 4200 November 14, 2002 NV28M 200 400 4:2:8:4 800 800 1600 100 6.4 8.1 1.3
  1. ^ Pixel shaders: vertex shaders: texture mapping units: render output units

GeForce4 Go driver support

[edit]

This family is a derivative of the GeForce4 MX family, produced for the laptop market. The GeForce4 Go family, performance wise, can be considered comparable to the MX line.

One possible solution to the lack of driver support for the Go family is the third party Omega Drivers. Using third party drivers can, among other things, invalidate warranties. The Omega Drivers are supported by neither laptop manufacturers, laptop ODMs, nor by Nvidia. Nvidia attempted legal action against a version of Omega Drivers that included the Nvidia logo.[30]

Support

[edit]

Nvidia has ceased driver support for GeForce 4 series.

Final drivers

[edit]
  • Windows 9x & Windows Me: 81.98 released on December 21, 2005; Download;
Product Support List Windows 95/98/Me – 81.98.
  • Driver version 81.98 for Windows 9x/Me was the last driver version ever released by Nvidia for these systems; no new official releases were later made for these systems.
  • Windows 2000, 32-bit Windows XP, 64-bit Windows XP & Media Center Edition: 93.71 released on November 2, 2006; Download.
  • Windows 2000, 32-bit Windows XP, 64-bit Windows XP & Media Center Edition: 93.81 (beta) released on November 28, 2006; Download.

The drivers for Windows 2000/XP can also be installed on later versions of Windows such as Windows Vista and 7; however, they do not support desktop compositing or the Aero effects of these operating systems.

Note: Despite claims in the documentation that 94.24 (released on May 17, 2006) supports the Geforce 4 series, it does not (94.24 actually supports only GeForce 6 and GeForce 7 series).[31]

(Products supported list also on this page)

Windows 95/98/Me Driver Archive
Windows XP/2000 Driver Archive

  • Linux/BSD/Solaris: 96.43.23 released on September 14, 2012; Download;

Unix Driver Archive

See also

[edit]

Notes and references

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

GeForce 4 series

View on Grokipedia
from Grokipedia
The GeForce 4 series is the fourth generation of NVIDIA's GeForce-branded graphics processing units (GPUs), released in early 2002 as a successor to the GeForce 3 lineup, offering enhanced 3D graphics performance for gaming and multimedia applications through innovations like the nFiniteFX II engine and support for DirectX 8.1. The series is divided into two primary sub-lines: the high-performance GeForce4 Ti models—the initial models (Ti 4600, Ti 4400, Ti 4200) built on the NV25 chip fabricated using a 150 nm TSMC process with 63 million transistors, and the later Ti 4800 built on the NV28 chip—and the more affordable mainstream GeForce4 MX models (MX 460, MX 440, MX 420, and MX 4000), based on the NV17 core. The Ti variants featured up to 128 MB of DDR memory on a 128-bit bus, core clocks ranging from 250 MHz (Ti 4200) to 300 MHz (Ti 4600), and integrated features such as 4 pixel shaders, 8 texture mapping units (TMUs), 4 render output units (ROPs), and hardware transform and lighting (T&L) capable of 136 million vertices per second. In contrast, the MX models targeted budget users with 64 MB of DDR or SDR memory on a narrower 128-bit or 64-bit bus, lower clocks (e.g., 250 MHz core for MX 4000), and reduced capabilities like 2 pixel shaders and 4 TMUs, prioritizing cost-efficiency over peak performance. Key innovations in the GeForce 4 series included the AccuView antialiasing technology, which delivered up to 4.8 billion samples per second on the Ti 4600 for smoother visuals, and the nView desktop management system supporting up to 16 displays for enhanced setups. Additionally, it incorporated a Video Processing Engine (VPE) with hardware decoding, , and an integrated TV encoder for improved video playback and output quality, marking a step forward in multimedia integration. Mobile variants under the GeForce4 Go branding, such as the Go 420 and Go 440, extended these features to laptops with power-optimized designs. Overall, the series provided a balance of performance gains—up to 10.4 GB/s on top models—and broader accessibility, influencing PC gaming during the early 2000s era of titles like .

Overview

Development and Release

The high-end GeForce 4 Ti series was developed under NVIDIA's codename (NV25), representing a direct revision and refinement of the GeForce 3's underlying design, while the MX models utilized a budget-oriented derivative of the architecture (NV17/NV18), allowing to target both premium enthusiasts and mainstream consumers with a unified family approach. NVIDIA announced the GeForce 4 series on February 6, 2002, positioning it as the direct successor to the GeForce 3 released the previous year, with initial availability for Ti and cards rolling out immediately thereafter. Concurrently, the mobile GeForce4 Go series launched to extend the lineup to laptops, featuring models like the 420 Go and 440 Go derived from the architecture. Later in the year, on November 14, 2002, NVIDIA added the GeForce4 4200 Go as an upgraded mobile option based on the NV28 processor. However, the series faced criticism for its limited performance gains over previous generations and absence of advanced support, leading to accusations of misleading branding. At launch, suggested retail pricing reflected the tiered strategy: the flagship GeForce4 Ti 4600 at $399, the GeForce4 Ti 4400 at $299, the GeForce4 Ti 4200 at $199 (released in April 2002), and the entry-level GeForce4 MX 440 at approximately $149. Production of the series wound down around 2004 as NVIDIA shifted focus to the succeeding GeForce FX lineup, though rebranded variants like the GeForce4 Ti 4800 SE—essentially an overclocked NV28-based refresh of the Ti 4800—were introduced in early to extend market presence.

Key Features and Innovations

The GeForce 4 series represented a significant step forward in NVIDIA's graphics technology, introducing enhancements in shader programmability, , , and that improved both gaming performance and multimedia functionality. Built on the architecture for Ti models and a derivative for MX, the series supported key industry standards, with the high-end Ti variants offering full compatibility with 8.1, including Pixel Shader 1.3 for advanced per-pixel effects like and Vertex Shader 1.1 for geometry transformations, alongside 1.3 for cross-platform rendering. Central to the Ti models was the nFiniteFX Engine II, which delivered up to three times the vertex processing power of the prior generation through dual programmable vertex s and optimized pixel shader pipelines, enabling complex effects such as procedural deformations, real-time fur shading, and fog simulation. Complementing this, Intellisample 2.0 provided advanced multisampling with modes like 4x and under the Accuview , achieving 2-3 times better performance in antialiased scenes compared to competitors while minimizing impacts. The Lightspeed Memory Architecture II (LMA II) further boosted efficiency with four independent 32-bit memory controllers, quad caching, and 4:1 lossless Z-compression for texture and depth data, resulting in up to three times faster memory access in demanding scenarios. TwinView supported seamless dual-monitor setups, allowing independent analog, digital, or TV outputs for enhanced productivity and immersion. In the budget-oriented MX variants, the Video Processing Engine (VPE) introduced hardware-accelerated iDCT (inverse ) and VLC (variable length code) decoding for video, marking a major upgrade in media handling with features like , adaptive de-interlacing, and 5x3-tap filtering for smooth playback and scaling. These cards also included integrated TV encoders for high-quality component output, a first for GPUs, which helped bridge high-performance gaming with consumer applications like DVD and HDTV viewing.

Architecture

Core Design and Kelvin Architecture

The GeForce 4 series employs two distinct GPU microarchitectures: the architecture for high-end Ti variants and a derivative of the Celsius architecture for budget MX variants. , NVIDIA's fourth-generation and successor to the third-generation Celsius, powers the NV25 chip in Ti models and was fabricated by using a 0.15 μm process. The NV25 core integrates 63 million transistors across a die size of 142 mm². By comparison, the NV17 core in the MX variants, based on a cost-reduced Celsius design, contains 29 million transistors on a smaller 65 mm² die, also at 0.15 μm, reflecting its streamlined design for cost-sensitive applications. Subsequent revisions such as the NV28 (for later Ti models like the Ti 4800) remained on the 0.15 μm process but enabled higher clock speeds through design optimizations. At its foundation, the Kelvin architecture utilizes a quad-pipe rendering setup in the Ti series, equipped with 4 pixel pipelines and 2 programmable vertex engines to handle and rasterization efficiently. The MX series, however, employs a reduced configuration with 2 pixel pipelines and omits programmable vertex shaders, prioritizing compatibility with legacy 7 features while sacrificing advanced programmable effects. This hardware layout supports the nFiniteFX Engine II for enhanced lighting and texturing capabilities in Ti models. Memory subsystems in the Ti models feature a unified 128-bit DDR controller, delivering bandwidth improvements over the GeForce 2's SDR implementation through better prefetching and tiling mechanisms. MX variants offer flexible 64/128-bit DDR controllers, allowing manufacturers to balance performance and cost by configuring narrower buses in entry-level configurations like the MX 440 SE. The Ti series, exemplified by the GeForce4 Ti 4600, operates with typical power consumption of around 50 W, drawn entirely from the motherboard via the standard AGP 4x interface without auxiliary connectors, enabling straightforward integration into contemporary PC systems.

Graphics Pipeline and Shaders

The GeForce 4 Ti series featured four parallel rendering pipelines, enabling efficient processing of graphics data through the Kelvin architecture's programmable shading capabilities. These pipelines supported Pixel Shader 1.3, allowing up to four instructions per pixel for advanced effects such as per-pixel lighting and , and Vertex Shader 1.1 for geometric transformations and of up to two bones per vertex. This setup marked a significant advancement in programmable rendering, facilitating 8.1 compliance and improved scene complexity handling compared to prior generations. In contrast, the GeForce 4 MX series lacked programmable entirely, relying on a fixed-function pipeline compliant with 7.0 standards for basic transform and lighting operations. This limitation restricted the MX variants to simpler rendering tasks without the extensibility for custom effects, though software emulation could approximate some Vertex Shader 1.1 functionality in compatible applications. Both the Ti and MX series incorporated four texture units to support bilinear filtering and multi-texturing operations, including in the Ti models for enhanced . However, the MX units operated without shader extensibility, capping their utility at fixed-function texture combinations. For antialiasing and , the Ti series employed Intellisample 2.0, an adaptive technique that combined multisampling with transparency to reduce edge while preserving performance. The MX series, meanwhile, supported only basic multisampling without advanced adaptive features. Conceptually, shader throughput in the Ti series could be estimated by the pixel fill rate, calculated as the core clock frequency multiplied by the number of pipelines; for the GeForce 4 Ti 4600 operating at 300 MHz, this yielded 1.2 Gpixels/s (300 \times 10^6 \times 4 / 10^9). This metric highlighted the architecture's capacity for high-volume pixel processing in shader-intensive workloads.

High-End Variants

GeForce4 Ti Series Lineup

The GeForce4 Ti series represented NVIDIA's high-end desktop graphics offerings within the 4 lineup, built primarily on the NV25 graphics processor with later models using the NV28 variant for improved manufacturing yields and AGP 8x compatibility, including Ti 4200 8x configurations. These cards targeted gamers and professionals seeking enhanced performance over the standard GeForce4 models, featuring 128-bit memory interfaces and support for 8.1 and 1.3. The series launched with the Ti 4600 and Ti 4400 models on February 6, 2002, followed by the Ti 4200 in April 2002, and concluded with the limited-release Ti 4800 in March 2003 and its OEM Ti 4800 SE variant in February 2003. All models utilized DDR memory and delivered theoretical pixel fill rates up to 1.2 Gpixels/s and texture fill rates up to 2.4 Gtexels/s, depending on the clock speeds.
ModelRelease DateCore ClockMemory ClockMemory ConfigurationInterfaceGPUNotes
GeForce4 Ti 4600February 6, 2002300 MHz325 MHz128 MB DDR, 128-bitAGP 4xNV25Flagship model with highest clocks in initial lineup.
GeForce4 Ti 4400February 6, 2002275 MHz275 MHz128 MB DDR, 128-bitAGP 4xNV25Mid-tier option balancing cost and performance.
GeForce4 Ti 4200April 16, 2002250 MHz222 MHzUp to 128 MB DDR, 128-bitAGP 4xNV25Entry-level Ti variant, with some configurations supporting 64 MB. MSRP $199 (announced February 2002).
GeForce4 Ti 4800March 15, 2003300 MHz325 MHz128 MB DDR, 128-bitAGP 8xNV28Limited retail release aimed at competing with emerging rivals.
GeForce4 Ti 4800 SEFebruary 16, 2003275 MHz275 MHz128 MB DDR, 128-bitAGP 8xNV28OEM-exclusive variant with reduced clocks for system integration.
</section_text>

GeForce4 Ti Performance and Benchmarks

The GeForce4 Ti series marked a substantial advancement over the GeForce3 in gaming performance, particularly in 8-era titles, with the Ti 4600 capable of up to 20-50% higher frame rates compared to the GeForce3 Ti 500. This edge was most evident in shader-intensive scenarios, where the Ti series' dual texture rendering pipelines and improved fill rates provided smoother gameplay without requiring extensive CPU intervention. In competition with contemporaries like the ATI 8500, the GeForce4 Ti held its own, often outperforming it by a slim margin in key tests. For instance, the Ti 4600 achieved roughly 15% higher frame rates than the 8500 in , especially in scenes leveraging programmable shaders, thanks to NVIDIA's optimized 8 implementation and the 128-bit memory bus that sustained data throughput for complex effects. Synthetic evaluations further underscored this capability; the Ti 4600 typically scored around 10,000 points in 3DMark 2001, establishing it as a benchmark leader for high-end 8 workloads at the time. Under full load, power consumption hovered near 50W, balancing performance with reasonable efficiency for mid-2000s systems. The Ti lineup, particularly the Ti 4200, exhibited impressive longevity into the DirectX 9 era, remaining suitable for mid-range gaming up to 2005 in titles like at moderate resolutions and settings. However, limitations emerged in demanding configurations; the series faltered with high-resolution (e.g., 4x AA at 1600x1200), where the 128-bit bus constrained , leading to drops of up to 50% compared to lower-res tests and highlighting a key bottleneck against emerging 256-bit competitors.

Budget Variants

GeForce4 MX Series Lineup

The GeForce4 MX series was introduced as NVIDIA's budget-oriented graphics solution within the GeForce4 lineup, targeting entry-level desktop systems with cost-effective performance for basic 2D and 3D applications. Launched on February 6, 2002, the series utilized the NV17 graphics processor fabricated on a 150 nm process with 29 million transistors, featuring a fixed-function compliant with 7.0 but lacking programmable shaders. The MX lineup included three primary models: the MX 420, MX 440, and the rarer MX 460, all supporting AGP 4x interfaces and offering memory configurations optimized for affordability. A later refresh, the MX 4000, was released in December 2003 based on the NV18 chip, featuring similar specifications but with options for 128 MB memory and a 64-bit bus in some implementations. Variants emerged in 2002, such as the MX 440-8x based on the NV18 chip revision, which added AGP 8x compatibility while maintaining similar core specifications.
ModelRelease DateCore ClockMemory ClockMemory SizeMemory TypeBus WidthFill Rate (Pixel)Chip
MX 420Feb 6, 2002250 MHz166 MHz64 MBSDR128-bit0.50 Gpixels/sNV17
MX 440Feb 6, 2002275 MHz200 MHz64 MBDDR128-bit0.55 Gpixels/sNV17
MX 460Feb 6, 2002300 MHz275 MHz64 MBDDR128-bit0.60 Gpixels/sNV17
MX 4000Dec 14, 2003250 MHz166 MHz64/128 MBDDR/SDR128/64-bit0.50 Gpixels/sNV18
Note: The MX 440 and MX 460 supported optional 128 MB configurations in select board partner implementations, though 64 MB was standard; lower-end variants of the MX 440 (e.g., MX 440-SE) operated at reduced clocks like 250 MHz core and used 64-bit bus options for further cost reduction. The NV18 revision, introduced mid-2002, enabled AGP 8x on models like the MX 440-8x without altering clock speeds or memory setups.

GeForce4 MX Performance and Applications

The GeForce4 MX series provided sufficient performance for entry-level gaming, handling 7 and 8 titles adequately at low to medium settings on resolutions such as 800x600. For instance, the MX 440 achieved playable frame rates in older games like , making it a viable option for budget gamers during its era. Despite these capabilities, the series had notable limitations due to the absence of programmable vertex and shaders, which caused visual artifacts and suboptimal rendering in 8+ applications. In gaming comparisons, the MX 440 was approximately 50% slower than the higher-end GeForce4 Ti 4200 overall, and significantly more in shader-dependent scenarios. Benchmark results underscored its mid-range positioning, with the MX 440 scoring approximately 4,500 points in 3DMark 2001 under typical configurations. It also excelled in 2D acceleration, benefiting office productivity tasks by enabling smooth high-resolution desktop rendering and window management. Beyond gaming, the GeForce4 MX shone in and professional applications, outperforming contemporary integrated graphics solutions. In CAD software like 2002, it delivered reliable 2D performance for design workflows, leveraging its robust rendering engine. The integrated Video Processing Engine (VPE) further enhanced video handling, supporting full hardware decoding—including inverse (iDCT) and variable-length coding (VLC)—for efficient playback in tools like , with minimal CPU overhead. The 128-bit contributed to these strengths by providing balanced bandwidth for tasks.

Mobile Graphics

GeForce4 Go Series

The GeForce4 Go series represented NVIDIA's initial foray into mobile graphics processing units (GPUs) under the architecture umbrella, tailored for integration with reduced power profiles and compact form factors compared to desktop counterparts. Announced in early 2002, these GPUs derived from the desktop GeForce4 MX and Ti architectures, incorporating features like integrated transform and lighting (T&L) engines while prioritizing portability. The lineup targeted entry-level to mid-range mobile gaming and multimedia applications, supporting 8.0 in MX-based models and 8.1 in the Ti-based 4200 Go variant. The series debuted with the entry-level GeForce4 420 Go and 440 Go on February 6, 2002, both utilizing the NV17M chip fabricated on a 150 nm process. The 420 Go operated at a 200 MHz core clock and 200 MHz memory clock (400 MHz effective) with 32 MB of DDR memory across a 64-bit bus, suitable for basic 3D acceleration in thin-and-light notebooks. In contrast, the 440 Go offered improved performance with a 220 MHz core clock and 220 MHz memory clock (440 MHz effective), paired with 64 MB DDR on a wider 128-bit bus, enabling better handling of textures and resolutions for mobile productivity and light gaming. Later releases expanded capabilities with the GeForce4 460 Go, launched on October 14, 2002, still based on the NV17M chip. It featured a higher 250 MHz core clock and 250 MHz memory clock (500 MHz effective) with 64 MB DDR on a 128-bit bus, providing enhanced fill rates for more demanding portable workloads. The series culminated in the GeForce4 4200 Go on November 14, 2002, shifting to the NV28M chip for Ti-level features including 4 pixel shaders and 2 vertex shaders under 8.1 support. This model ran at a 200 MHz core clock but with a 200 MHz memory clock (400 MHz effective), 64 MB DDR, and 64-bit bus, balancing efficiency and performance in premium laptops. These GPUs were typically implemented in either soldered configurations directly onto the for cost-effective designs or as early (MXM) cards for easier upgrades in select high-end systems.
ModelRelease DateCore ClockMemory Clock (Effective)Memory Size/TypeBus WidthChip
Go 420Feb 6, 2002200 MHz200 MHz (400 MHz)32 MB DDR64-bitNV17M
Go 440Feb 6, 2002220 MHz220 MHz (440 MHz)64 MB DDR128-bitNV17M
Go 460Oct 14, 2002250 MHz250 MHz (500 MHz)64 MB DDR128-bitNV17M
Go 4200Nov 14, 2002200 MHz200 MHz (400 MHz)64 MB DDR64-bitNV28M

Power and Thermal Considerations

The GeForce4 Go series was engineered with laptop-specific adaptations to manage power draw and heat generation, prioritizing battery life and thermal stability in portable systems. Early models, such as the GeForce4 420 Go and 440 Go based on the architecture, typically consumed 10-15W under load, enabling longer battery runtime but capping performance for lighter tasks like office applications and basic multimedia. These variants benefited from NVIDIA's PowerMizer technology, which dynamically adjusted clock speeds to reduce use by up to 25%. Higher-end models like the GeForce4 Go 4200, built on the more advanced NV28 architecture, exhibited a higher TDP of approximately 25W, resulting in greater heat output that frequently caused thermal throttling in slim chassis and demanded robust solutions such as heat pipes and dual fans. This lack of advanced power-saving circuitry—unlike in MX-based predecessors—exacerbated battery drain and thermal challenges, making it less ideal for ultrathin designs without enhanced ventilation. For instance, in 2002 models like the 8200 and 5100 equipped with the Go 4200, reported overheating prompted BIOS-level performance limits to enforce safer operating temperatures and avert hardware damage. Power efficiency varied significantly across the lineup, with MX-derived Go chips achieving idle consumption around 2W, far superior to Ti-based variants that lacked dynamic clocking and drew more even at rest. To optimize for mobile use, NVIDIA implemented lower operating voltages between 1.5V and 1.8V across the series, while select models integrated TV-out capabilities directly on the GPU to eliminate separate chips and further curb power needs. Higher-spec Go configurations, including the 4200, employed a 64-bit memory bus for improved bandwidth efficiency in power-constrained environments, though this contributed to their elevated thermal profile in constrained laptop environments.

Software and Driver Support

Driver Releases and Compatibility

The GeForce 4 series launched with initial support through NVIDIA's drivers in March 2002, compatible with and XP operating systems. Key subsequent releases included the ForceWare 81.98 drivers on December 21, 2005, which marked the final official update for and Me systems and provided broad compatibility for legacy hardware. The 93.71 drivers, released on November 2, 2006, served as the last major update for and XP, incorporating optimizations for the series' hardware features. For Linux users, NVIDIA's final dedicated driver for the series was version 96.43.23, released on September 14, 2012, offering 2D and 3D acceleration for 4 variants including the Ti and MX lines. Compatibility across the lineup varied by model: the high-end Ti and Go 4200 variants delivered full hardware support for 8.1, enabling advanced and vertex effects. In contrast, the budget MX series was hardware-limited to 7, relying on software emulation for partial 8 functionality in compatible applications. NVIDIA's driver ecosystem for the GeForce 4 series utilized a unified up to the R177 branch, streamlining updates across GeForce generations from the 4 series onward. Unofficial third-party options, such as the Omega Drivers, extended compatibility and tweaks for the series into 2008, though they lacked certification and were developed independently.

Discontinued Support and Legacy Use

NVIDIA ended driver support for the GeForce 4 series on Windows operating systems with the release of ForceWare version 93.71 on November 2, 2006, which provided compatibility for and Windows 2000. For , support concluded with the 96.43.xx legacy driver branch, culminating in version 96.43.23 released on September 14, 2012, after which no further updates were issued. These final drivers mark the endpoints of maintenance, leaving the hardware exposed to vulnerabilities from evolving operating system architectures and security requirements without patches. Legacy drivers remain available for download from NVIDIA's archives as of 2025. The GeForce 4 series is incompatible with key features in and later, such as the Aero interface, due to its reliance on the older XPDM driver model rather than the required WDDM standard. Operation on and subsequent versions demands unofficial modifications, like installations of XP-era drivers, which often result in instability or disabled visual effects. For optimal performance, users must revert to legacy environments like or Vista, where the original drivers enable full functionality without such workarounds. As of 2025, the GeForce 4 series finds primary use in retro gaming setups configured with period-accurate hardware, such as 4-based systems running or XP to authentically experience early 2000s titles like or . These cards also integrate well with emulation tools like on modern hosts for running DOS-era games, preserving hardware authenticity without needing contemporary GPU acceleration. Community-driven modifications, including custom driver tweaks and NvAPI compatibility wrappers, allow limited integration into newer systems for niche testing or preservation projects, though performance remains constrained by the card's age. Beyond gaming, the series holds value as collectibles among hardware enthusiasts, with models like the GeForce4 Ti 4600 prized for their historical significance in NVIDIA's evolution toward programmable shading. In select industrial contexts, GeForce4 MX variants persist in legacy embedded systems or stable control applications, where their mature driver ecosystem ensures reliable operation without the overhead of modern features. The absence of updates since 2012 heightens risks from OS kernel changes or deprecated APIs, underscoring the need for isolated, air-gapped environments to maintain viability.

Market Impact and Comparisons

Competition with ATI Radeon Series

The GeForce 4 Ti 4600 provided strong competition to ATI's 8500 (R200) in 8 workloads, often matching or exceeding its performance in rasterization-heavy scenarios. In benchmarks such as at 1600x1200 resolution, the Ti 4600 achieved roughly double the of the Radeon 8500, demonstrating superior fill rate and texture handling capabilities. However, the Radeon 8500 held an edge in vertex shading due to its dual programmable vertex shaders, which offered greater flexibility for complex geometry processing compared to the GeForce 4 Ti's fixed-function enhancements. The GeForce 4 Ti series also led in anti-aliasing efficiency, with features like multisampling delivering playable performance at 2x FSAA in titles like , where the Radeon 8500's approach incurred higher performance penalties. In the budget segment, the GeForce 4 MX 440 competed closely with the 9000 (RV250) on price, both targeting around $100, but the offered partial 8.1 support including basic and vertex shaders, while the MX 440 was 8.0 compliant but lacked programmable shaders, relying on fixed-function capabilities similar to the older GeForce 2 architecture. Performance was mixed, with the MX 440 edging out in some 7/8 fixed-function games like Quake III due to its optimized drivers, but the 9000 pulled ahead in newer titles and provided better scalability. This positioned the 9000 as a more future-proof option for entry-level users transitioning to shader-based effects. NVIDIA maintained a dominant market position in 2002, capturing approximately 63% of the U.S. discrete GPU unit share amid the 4 launch, bolstered by strong OEM adoption and driver stability. ATI began gaining traction in niche areas like , where its drivers achieved initial 3D acceleration support by late 2002, appealing to open-source enthusiasts despite NVIDIA's earlier proprietary binaries. Key differentiators included NVIDIA's TwinView for seamless dual-monitor setups with independent resolutions and NVIDIA's nView desktop management, versus ATI's HydraVision, which emphasized virtual display partitioning and bezel compensation for productivity workflows on cards. In video processing, the series excelled in hardware-accelerated decoding and for DVD playback, outperforming 4 models in image quality for media center applications. The later-launching 9700 (R300) in August 2002 shifted the balance decisively against the GeForce 4 Ti 4400, with early benchmarks showing the delivering up to 175% higher scores in 8.1 titles like 2001 (e.g., 7800 for 9700 Pro vs. 2835 for Ti 4600), particularly when enabling anti-aliasing or anisotropic filtering. In transform and lighting tests, the 9700 achieved scores three times those of the Ti 4400, highlighting its eight pixel shader processors versus the GeForce's four. While the Ti 4400 remained competitive in legacy 8 games without advanced effects—trailing by about 10-20% in some rasterization benchmarks—the 9700's full 9 compliance and superior shader performance cemented ATI's lead in high-end positioning by mid-2002.

Reception and Long-Term Legacy

The GeForce 4 Ti series was widely acclaimed upon release for its architectural advancements, including enhanced pixel shader support up to version 1.3 and improved multisampling , which delivered superior image quality and performance in 8-era games compared to predecessors like the GeForce 3. Reviewers highlighted the Ti 4600 as a standout, praising its high of up to 10.4 GB/s and ability to handle demanding titles at high resolutions, positioning it as a premium choice for . In contrast, the GeForce4 MX lineup drew sharp criticism for its deceptive "GeForce 4" branding, as it relied on the older NV17 core derived from the GeForce 2 architecture, lacking key innovations like vertex shaders and offering performance closer to mid-range cards from prior generations. This naming strategy confused consumers and enthusiasts, leading to backlash over perceived , though it succeeded in capturing budget market share. Commercially, the GeForce 4 series bolstered NVIDIA's position as the leading GPU vendor in the early , with the company's graphics revenue surging 39% to $1.91 billion in 2003 amid strong demand for both and variants. This success helped NVIDIA maintain dominance over competitors like ATI ahead of the GeForce FX transition, solidifying its hold on the discrete graphics market during a period of rapid industry growth. In terms of long-term legacy, the GeForce 4 series marked a pivotal step in mobile graphics with the introduction of the Go lineup, such as the , which brought desktop-level 8 features to laptops for the first time on NVIDIA's 150 nm process. It also advanced technology by extending capabilities to version 1.3, laying groundwork for the more programmable shaders in the subsequent and facilitating the broader industry shift toward 8 compliance. By 2025, the series retains significant retro appeal among enthusiasts for emulation and vintage PC builds, particularly the Ti 4600, which excels in running early 2000s software on legacy operating systems like due to its robust 8 support. Culturally, it epitomized high-end gaming PCs of the era, powering iconic setups in titles like , and commands collectible values of $50–$200 on secondary markets for well-preserved units. Official driver support for the series concluded with ForceWare release 93.71 in 2006.

References

  1. https://www.eweek.com/it-management/nvidia-announces-geforce4/
  2. https://www.techpowerup.com/gpu-specs/geforce4-ti-4600.c178
  3. https://www.techpowerup.com/gpu-specs/geforce4-mx-4000.c776
  4. https://www.techpowerup.com/gpu-specs/geforce4-mx-440.c781
  5. https://www.techpowerup.com/gpu-specs/geforce4-420-go.c1553
  6. https://www.techpowerup.com/gpu-specs/geforce4-go-4200.c1558
  7. https://www.techpowerup.com/gpu-specs/geforce4-ti-4200.c2134
  8. https://www.vgamuseum.info/images/doc/nvidia/gf4ti_overview.pdf
  9. https://www.beyond3d.com/content/reviews/20/4
  10. https://www.techpowerup.com/gpu-specs/nvidia-nv25.g23
  11. https://www.techpowerup.com/gpu-specs/nvidia-nv17.g632
  12. https://www.techpowerup.com/gpu-specs/nvidia-nv28.g650
  13. https://old.hotchips.org/wp-content/uploads/hc_archives/hc14/3_Tue/18_moreton.pdf
  14. http://ixbtlabs.com/articles/gf4mx/index.html
  15. http://ixbtlabs.com/articles/gf4/index2.html
  16. https://www.evga.com/products/specs/gpu.aspx?pn=4D8D1178-BCB2-4C1F-B82C-0D5199EB3BEC
  17. https://thg.ru/graphic/20020206/
  18. https://en.gamegpu.com/geForce-4-series/geforce-4-ti-4200
  19. https://developer.valvesoftware.com/wiki/DirectX_Versions
  20. https://forums.guru3d.com/threads/opengl-d3d.172037/
  21. https://www.geocities.ws/ng_zz/oomz/GeF4-b.html
  22. https://www.philscomputerlab.com/geforce4-mx-460.html
  23. http://ixbtlabs.com/articles/gf4/index1.html
  24. https://www.techpowerup.com/gpu-specs/geforce4-ti-4200-8x.c182
  25. https://www.nvidia.com/en-us/drivers/det-xp-win9x-2942/
  26. https://www.techpowerup.com/gpu-specs/geforce4-ti-4400.c180
  27. https://www.techpowerup.com/gpu-specs/geforce4-ti-4800.c2139
  28. https://www.techpowerup.com/gpu-specs/geforce4-ti-4800-se.c179
  29. https://www.pugetsystems.com/labs/articles/nvidia-geforce4-how-does-it-compare-5/
  30. https://forums.anandtech.com/threads/geforce-3-ti-200-or-geforce-4-ti4200.775093/
  31. https://www.activewin.com/reviews/hardware/graphics/nvidia/gf4ti4600/benchs2.shtml
  32. https://www.pcstats.com/articles/1113/10.html
  33. https://forums.anandtech.com/threads/ti4200-power-usage.1057236/
  34. https://www.vogons.org/viewtopic.php?t=52495
  35. https://www.techpowerup.com/gpu-specs/geforce4-mx-440-8x.c783
  36. https://www.techpowerup.com/gpu-specs/geforce4-mx-420.c777
  37. https://www.techpowerup.com/gpu-specs/geforce4-mx-460.c786
  38. https://www.techpowerup.com/gpu-specs/geforce4-mx-440-se.c780
  39. https://www.youtube.com/watch?v=cJ1F6cZDJdQ
  40. https://forums.whirlpool.net.au/archive/71086
  41. https://www.512bit.net/3dmark2001se.html
  42. https://bjorn3d.com/2002/05/xfxs-geforce4-mx-series/
  43. https://forums.anandtech.com/threads/3dlabs-oxygen-gmx-2000-96mb-vs-geforce4-mx-440-for-2d-autocad.796432/
  44. https://forum.doom9.org/archive/index.php/t-125376.html
  45. https://hothardware.com/reviews/nvidia-geforce-4-mx440-roundup
  46. https://hitesti.com/compare/gpu/geforce4-420-go-vs-nvidia-geforce-mx330
  47. https://technical.city/en/video/Radeon-9100-vs-GeForce4-420-Go
  48. https://thg.ru/mobile/20020408/
  49. https://www.dell.com/community/Laptops-General-Read-Only/Inspiron-5100-alternate-fixes-to-overheating/td-p/1044922
  50. https://www.guru3d.com/download/detonator-28-32-b01a/
  51. https://www.nvidia.com/en-us/geforce/drivers/results/11/
  52. https://www.nvidia.com/en-us/drivers/details/48996/
  53. https://nvidia-omega-drivers.updatestar.com/en
  54. https://www.nvidia.com/en-us/drivers/results/11/
  55. https://www.tomshardware.com/reviews/pc-graphics-xbox,423-11.html
  56. https://hothardware.com/reviews/128mb-radeon-8500-les-vs-nvidias-geforce4-ti-4200-
  57. https://www.techpowerup.com/gpu-specs/radeon-9000-pro.c100
  58. https://www.ign.com/articles/2002/07/22/nvidia-selling-like-mad
  59. https://m.hexus.net/tech/reviews/graphics/18017-ati-vs-nvidia-linux-showdown/
  60. https://www.ign.com/articles/2002/04/02/ati-radeon-8500-64mb-review
  61. https://www.zdnet.com/article/radeon-9700-pro-benchmark-results/
  62. https://www.ign.com/articles/2002/04/23/evga-e-geforce4-mx-440-review
  63. https://www.osnews.com/story/649/john-carmack-dont-buy-a-geforce4-mx-for-doom-3/
  64. https://www.sec.gov/Archives/edgar/data/1045810/000104581004000007/annualreport.htm
  65. https://www.theregister.com/2003/08/08/nvidia_q2_sales_income_up1/
  66. https://www.vogons.org/viewtopic.php?t=44683
  67. https://www.nvidia.com/en-us/drivers/details/827/
Add your contribution
Related Hubs
User Avatar
No comments yet.