GeForce 4

A GeForce 4 is a fourth-generation graphics processing unit (GPU) manufactured by Nvidia which forms the basis of many computer graphics cards. Strictly speaking, the GeForce 4 is the chip, not the entire card, but in common usage this distinction tends to be ignored.

There are two different GeForce 4 families, the high-performance Ti family, and the budget MX family. Each of the two has been made in several different variants.

Contents

1 Key 2 GeForce 4 Ti 3 GeForce TI chipset table 4 GeForce 4 MX 5 GeForce MX chipset table 6 External links

Key

• NV17 - GeForce 4 MX (AGP-4X)
• NV18 - GeForce 4 MX (AGP-8X)
• NV19 - GeForce PCX (PCI Express)
• NV25 - GeForce 4 Ti (AGP-4X)
• NV28 - GeForce 4 Ti (AGP-8X)

GeForce 4 Ti

The GeForce 4 Ti (NV25) was launched in April 2002. It was very similar to the GeForce 3; the main differences were higher core and memory speeds, a revised memory controller, improved vertex and pixel shaders, hardware anti-aliasing and DVD playback. Proper dual-monitor support was also brought over from the GeForce 2 MX. The GeForce 4 Ti outperformed the older GeForce 3 by a significant but modest margin.

The initial two models were the Ti4400 and the top-of-the-range Ti4600. At the time of their introduction, Nvidia's main products were the entry-level GeForce 2 MX, the older but still high-performance GeForce 3 (now demoted to the upper mid-range), and the GeForce 4 models. However, ATI's Radeon 8500LE was somewhat cheaper than the Ti4400, and outperformed its price competitor, the GeForce 3 Ti200. In consequence, Nvidia introduced a new, slightly cheaper model: the Ti4200. In an attempt to prevent the Ti4200 damaging the Ti4400's sales, Nvidia set the Ti4200's memory speed at 444MHz on the models with a 128MiB frame buffer - a full 106MHz slower than the Ti4400 (all of which had 128MiB frame buffers). Models with a 64MiB frame buffer were set to 500MHz 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). Furthermore, many graphics card makers simply ignored Nvidia's guidelines, and set the memory speed at 500MHz on the 128MiB models anyway.

In late 2002 the NV25 core was replaced by the NV28 core, which differed only by addition of AGP-8X support. The NV28 based Ti4200s all had their memory set at 500MHz regardless of frame buffer size, but Nvidia rather surprisingly chose to make a Ti4400 equivalent - the Ti4800SE. The top-end NV28 core was the Ti4800, which in spite of a name that would suggest higher performance than the Ti4600, was clocked identically.

Performance-wise, all the GeForce 4 Ti chips were faster than GeForce 3 or Radeon 8500 based chips. The Ti4200 remained the best balance between price and performance until the launch of the Radeon 9500 Pro later in the year, and the Ti4600 generally got beaten by the Radeon 9700, but maintained an advantage in OpenGL software.

GeForce TI chipset table

NOTE: These are the official specifications dictated by NVIDIA; in practice the speeds tended to vary. All GeForce 4 Ti chips use a 128-bit memory bus. Table is slowest to fastest.

GeForce4 Chipset	Core	Core Speed (MHz)	Memory Speed (MHz)
Ti4200	NV25	250	500/444
Ti4200-8X	NV28	250	500/514
Ti4400	NV25	275	550
Ti4800SE	NV28	275	550
Ti4600	NV25	300	650
Ti4800	NV28	300	650

GeForce 4 MX

If the capabilities of the Geforce4 family are defined by the Geforce 4 Ti, then the GeForce 4 MX (NV17) is a Geforce4 in name only. On its release, disappointed enthusiasts described the Geforce 4 MX as a GeForce 2 MX with a better (128-bit DDR) memory controller. The GeForce 4 MX lacked vertex or pixel shaders of its bigger brother Geforce 4 Ti, and consequently was a distinctly lack-luster platform for gaming. But it also owed a good deal 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 GeForce 4 MX returned results within a single-digit percentage of Ti cards six or seven times the price.)

Despite harsh criticism by gaming enthusiasts, the Geforce 4 MX was a market success. Priced about 30% above the GeForce 2 MX, it provided marginally better performance, the ability to play (however slowly) a number of popular games that the GeForce 2 was not compatible with and—above all else—to the average non-specialist it sounded as if it was a "real" GeForce 4—i.e., a GeForce 4 Ti. Although it was comprehensively out-performed by the older and more expensive GeForce 3, many buyers were caught unawares. It was particularly successful in the PC OEM market, and rapidly replaced the GeForce 2 MX as the best-selling GPU.

There were 3 initial models - the MX420, the MX440 and the MX460. The MX420 was designed for very low end PCs, and the MX440 was a mass-market OEM solution. However, the MX460 was priced not far below the GeForce 4 Ti4200, the GeForce 3 Ti200 and the Radeon 8500LE/9100 (even the full 8500 in some cases), all of which outpeformed it easily. The end result was that the MX460 never had anywhere to go in the market, and flopped.

In terms of 3D-performance, the MX420 performed only slightly better than the GeForce2 MX400 and below the GeForce2 GTS, but 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 practise however, its main competitors were actually chipset-integrated graphics solutions, such as Intel's 845G and NVIDIA's own nForce 2.

The MX440 performed reasonably well for its intended audience, outperforming its closest competitor, the ATI Radeon 7500, as well as the older GeForce2 Ti. The MX460, while not slow by any means, simply didn't peform well enough compared to the Ti4200/Ti200/8500LE. 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 Vertex and Pixel shaders. Nvidia's answer to the ATI Radeon 9000 was the GeForce FX 5200.

In motion-video applications, the GeForce4 MX did pull ahead of its GeForce siblings. The GeForce4 MX (and not the GeForce4 Ti) introduced a significantly upgraded video-engine, the VPE (video processing engine.) VPE was a major upgrade from previous GeForce processors, containing more hardware to assist MPEG-2 playback. The GeForce4 MX was the first GeForce to offer hardware-iDCT and VLC (variable length code) decoding. With VPE, NVIDIA could finally compete head-to-head with ATI's outstanding video-engine.

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 was never updated; in fact, it had been discontinued several months previously. Another variant followed in late 2003 - the MX 4000, which was a GeForce 4 MX440SE with a slightly higher memory clock.

Surprisingly, the GeForce 4 MX line received a third update in 2004, with the PCX 4300 - an MX 4000 with support for PCI Express, and a wider memory bus. In spite of its new codename (NV19), the PCX 4300 is in fact simply an NV18 core with a chip bridging the NV18's native AGP interface with the PCI-Express bus.

GeForce MX chipset table

NOTE: These are the official specifications dictated by NVIDIA; in practice the speeds tended to vary. Table is slowest to fastest.

GeForce4 Chipset	Core	Core Speed (MHz)	Memory Speed (MHz)
MX420	NV17	250	166 (128-bit)
MX440SE	NV18	270	333 (64-bit)
MX 4000	NV18	275	400 (64-bit)
PCX 4300	NV19	275	333 (128-bit)
MX440	NV17	270	400 (128-bit)
MX440-8X	NV18	275	500/511 (128-bit)
MX460	NV17	300	550 (128-bit)

External links

• NVIDIA: GeForce4 family of GPUs (http://www.nvidia.com/page/geforce4.html)

Template:NVIDIA