• Introducing XDA Computing: Discussion zones for Hardware, Software, and more!    Check it out!

Complete Guide to NVIDIA: Everything You Need To Know

Search This thread

kunalneo

Member
Nov 29, 2013
38
36
NVIDIA continues its dominance in the graphics card industry. The company’s primary GPU lineup under the GeForce brand has been around for over two decades with close to twenty iterations. The series includes discrete graphics processors for desktops and laptops. Fun fact- the name GeForce originally stood for "Geometry Force" since GeForce 256 was the first GPU for personal computers to calculate the transform-and-lighting geometry.


NVIDIA GeForce generations

Here’s a look at NVIDIA’s GeForce lineup:


GeForce 256

The first GeForce GPU in the lineup, the GeForce 256 (NV10) was launched in September 1999 and was the first consumer-level PC graphics chip that shipped with hardware transform, lighting, and shading.

GeForce 2 series

The following year NVIDIA launched the GeForce2 (NV15) that introduced a twin texture processor per pipeline (4x2) design, doubling texture fillrate per clock compared to GeForce 256. This was followed by the GeForce2 MX (NV11), which offered performance similar to the GeForce 256 but at a reduced cost.

GeForce 3 series

In 2001 NVIDIA launched the GeForce3 (NV20) which introduced programmable vertex and pixel shaders. A version of GeForce 3 codenamed NV2A was developed for the Microsoft Xbox game console.

GeForce 4 series

In February 2002 the GeForce4 Ti (NV25) was launched as a refinement to the GeForce3. It included enhancements to anti-aliasing capabilities, an improved memory controller, a second vertex shader, and a manufacturing process size reduction to increase clock speeds. The GeForce4 MX was also introduced as a budget option based on the GeForce2, with the addition of some features from the GeForce4 Ti.

GeForce FX series

The GeForce FX (NV30) introduced a big change in architecture. It brought support for the new Shader Model 2 specification and carried the 5000 model number, as it was the fifth generation of the GeForce family. The series was also infamous for its heating and noisy fan issues.

GeForce 6 series

Launched in April 2004, the GeForce 6 (NV40) added Shader Model 3.0 and fixed the weak floating point shader performance of its predecessor. It additionally implemented high-dynamic-range imaging, SLI (Scalable Link Interface), and PureVideo capability (integrated partial hardware MPEG-2, VC-1, Windows Media Video, and H.264 decoding and fully accelerated video post-processing).

GeForce 7 series

The GeForce 7 series (G70/NV47) was introduced in June 2005 and was the last NVIDIA GPU series to support the AGP bus. It offered a wider pipeline and an increase in clock speed along with new transparency supersampling and transparency multisampling anti-aliasing modes (TSAA and TMAA). A version of the 7950 GT, called the RSX 'Reality Synthesizer', was used as the primary GPU on the Sony PlayStation 3.

GeForce 8 series

The first GeForce (G80) to fully support Direct3D 10, the 8th-gen GeForce series was launched in 2006. It was made using a 90nm process and built around the new Tesla microarchitecture. It was eventually refined and the die size was shrunk down to 65nm. The revised design codenamed G92 was implemented into the 8 series with the 8800GS, 8800GT, and 8800GTS-512, and was launched in 2007.

GeForce 9 series

Revisions for the GeForce 8 series were introduced after a short period in 2008 where the 9800GX2 used two G92 GPUs, in a dual PCB configuration with a single PCI-Express 16x slot. It also included two separate 256-bit memory busses, one for each GPU and a total of 1GB of memory on the card. Later the 9800GTX was launched with a single G92 GPU, 256-bit data bus, and 512 MB of GDDR3 memory.

GeForce 100 series

The following year NVIDIA launched the GeForce 100 series which were essentially rebranded versions of the GeForce 9 series available only for OEMs, although the GTS 150 was briefly available to consumers.

GeForce 200

The GeForce 200 series included the GT200 65nm-based graphics processor that had a total of 1.4 billion transistors and was introduced in 2008. It was also the year when NVIDIA changed its card-naming scheme by replacing the series number with the GTX or GTS suffix and then adding model numbers after that. The series features the new GT200 core on a 65nm die. The GeForce GTX 260 and the GTX 280 were the first products in the series while the GeForce 310 was released in November 2009 as a rebrand of GeForce 210.

GeForce 300 series

The 300 series cards were launched during the same year and were pretty much rebranded versions of the 200 series with support for DirectX 10.1 and based on the newer Fermi architecture. These were limited to OEMs only.

GeForce 400 series

The GeForce 400 series, codenamed GF100 was introduced in 2010 based on the Fermi architecture. They were the first NVIDIA GPUs to utilize 1GB or more of GDDR5 memory. The GTX 470 and GTX 480 were criticized for their high power use, high temperatures, and loud noise. At the same time, the GTX 480 was the fastest DirectX 11 card.

GeForce 500 series

To fix the issues, NVIDIA brought the 500 series with a new flagship (GTX 580) GPU based on an enhanced GF100 architecture (GF110). It offered higher performance, less power utilization, heat, and noise than the preceding GTX 480. Additionally, the GTX 590 was also introduced that packed two GF110 GPUs on a single card.

GeForce 600 series

In 2010, NVIDIA announced the Kepler microarchitecture, manufactured with the TSMC 28nm fabrication process. The company started supplying their top-end GK110 cores for use in Oak Ridge National Laboratory's Titan supercomputer, leading to a shortage of GK110 cores. Eventually, NVIDIA had to use the GK104 core, which was originally intended for the mid-range segment, to power their flagship, the GTX 680. It was followed by the dual-GK104 GTX 690 and the GTX 670.

GeForce 700 series

In May 2013, NVIDIA announced the 700 series based on the Kepler architecture, although it finally featured a GK110 chipset-based card at the top of the lineup. The GTX 780 was a cut-down version of the GTX Titan that achieved nearly the same performance for two-thirds of the price. A week after the release of the GTX 780, NVIDIA announced the GTX 770 to be a rebrand of the GTX 680. It was followed by the GTX 760 which was also based on the GK104 core and similar to the GTX 660 Ti.

GeForce 800M series

The GeForce 800M series included rebranded 700M series parts based on the Kepler architecture and some lower-end parts based on the newer Maxwell architecture.

GeForce 900 series

In March 2013, NVIDIA announced the new Maxwell microarchitecture. It was released in September 2014 on the GeForce 900 series and was the last series to support analog video output through DVI-I.

GeForce 10 series

In March 2014, NVIDIA announced that the successor to Maxwell would be the Pascal microarchitecture and was finally introduced on the GeForce 10 series in May 2016. It included 128 CUDA cores per streaming multiprocessor, GDDR5X memory, unified memory, and NVLink.

GeForce 20 series

In August 2018, NVIDIA announced Turing architecture as a successor to Pascal. The new microarchitecture was made to accelerate the real-time ray tracing support and AI Inferencing. It included a new Ray Tracing unit (RT Core) which dedicated processors to the ray tracing in hardware. It also supported the DXR extension in Microsoft DirectX 12. The company also introduced DLSS (Deep Learning Super Sampling), a new form of anti-aliasing that used AI to provide sharper imagery with less impact on performance.

The first GPUs to utilize the architecture were primarily aimed at high-end professionals and were introduced under the Quadro series. Eventually, the GeForce RTX series with RTX 2080 Ti, 2080, and 2070 were announced in 2018 followed by the RTX 2060 in January 2019.

In July 2019, NVIDIA announced the GeForce RTX Super line of cards, a refresh of the RTX 20 series which featured higher-spec versions of the RTX 2060, 2070, and 2080.

GeForce 16 series

In February 2019, NVIDIA announced the GeForce 16 series. Based on the same Turing architecture used in the GeForce 20 series, this series omitted the Tensor (AI) and RT (ray tracing) cores. This series continues to offer a more affordable graphics solution for gamers while still attaining a higher performance compared to respective cards of the previous GeForce generations. Similar to the RTX Super refresh, NVIDIA announced the GTX 1650 Super and 1660 Super cards, in October 2019.

GeForce 30 series

The latest and the most powerful graphics cards from NVIDIA, the new 30-series take over from the 20-series and were announced in 2020. It introduced a massive jump over the predecessor and an excellent price to performance ratio. However, getting your hands on one is a difficult task.

Mobile GPUs

NVIDIA produced a wide range of graphics cards for notebooks as far as the GeForce 2 series, under the GeForce Go branding. Most of the features present in the desktop counterparts were made available on the mobile version. With the introduction of the GeForce 8 series, the GeForce Go brand was discontinued and mobile GPUs were now a part of the main GeForce GPUs, with an M suffix. Once again NVIDIA brought some changes and dropped the M suffix in 2016 with the launch of the laptop GeForce 10 series in an attempt to unify the branding between their desktop and laptop GPU offerings. Currently, the RTX 20, GTX 16 and RTX 30 series of GPUs are available as both desktop and laptop variants. NVIDIA also has the GeForce MX range of mobile GPUs intended for lightweight notebooks with entry-level performance.

Nomenclature

Ever since the launch of the GeForce 100 series NVIDIA has been using the following naming scheme for its products:

G, GT, No Prefix - Mostly user for entry-level category of graphics cards with the last two numbers ranging from 00 to 45. Example - GeForce GT 730, GeForce GT 1030

GTS, GTX, RTX - Mid-range category of graphics cards with the last two numbers ranging from 50 to 65. Example - GeForce GTX 1060, GeForce RTX 2060

GTX, RTX - High-end range of graphics cards with the last two numbers ranging from 70-95. Example - GeForce GTX 1080Ti, GeForce RTX 3090

NVIDIA also uses the ‘Super’ or ‘Ti’ suffixes for its graphics cards to signify incremental updates.
 

ahfdee

Member
Feb 18, 2021
11
2
I am having some issues with my GeForce GTX 1050Ti
My motherboard sometimes doesn't detect the card and I am forced to use the Intel inbuild GPU
 
My RAM has no issues
Any other solutions
i sent back a pc for a bad ram stick, but it was really hard to find, passed all dianostics, blue screens with all different errors, found the bad stick by taking one out and running for a while, did great, then swapped sticks out, and wouldnt boot, shipped the next day(i had already contacted the seller and got the return approved). but i got a better system for pretty much the same price
 

ahfdee

Member
Feb 18, 2021
11
2
i sent back a pc for a bad ram stick, but it was really hard to find, passed all dianostics, blue screens with all different errors, found the bad stick by taking one out and running for a while, did great, then swapped sticks out, and wouldnt boot, shipped the next day(i had already contacted the seller and got the return approved). but i got a better system for pretty much the same price
I think my GPU has Thermal issues
When I boot from the GPU's HDMI slot after 3 to 4 days of not using the PC the GPU works. Do a need to put some thermal paste?
 

ahfdee

Member
Feb 18, 2021
11
2
i'm no expert but if you had thermal issues, they wouldn't show after 3 or 4 days idle, and get msi afterburner and watch temps while using, for thermal shutdown you'd need to be at 100C or higher. Are you on winblows or linux?
Either way do a clean install of drivers and reset nvidia settings. How do you boot from an hdmi slot?
I can either boot from Nvidia HDMI slot or default Intel HDMI slot.
To boot from either of the slots i just take out the HDMI cable from one slot and to it into another one
 

Top Liked Posts

  • There are no posts matching your filters.
  • 3
    NVIDIA continues its dominance in the graphics card industry. The company’s primary GPU lineup under the GeForce brand has been around for over two decades with close to twenty iterations. The series includes discrete graphics processors for desktops and laptops. Fun fact- the name GeForce originally stood for "Geometry Force" since GeForce 256 was the first GPU for personal computers to calculate the transform-and-lighting geometry.


    NVIDIA GeForce generations

    Here’s a look at NVIDIA’s GeForce lineup:


    GeForce 256

    The first GeForce GPU in the lineup, the GeForce 256 (NV10) was launched in September 1999 and was the first consumer-level PC graphics chip that shipped with hardware transform, lighting, and shading.

    GeForce 2 series

    The following year NVIDIA launched the GeForce2 (NV15) that introduced a twin texture processor per pipeline (4x2) design, doubling texture fillrate per clock compared to GeForce 256. This was followed by the GeForce2 MX (NV11), which offered performance similar to the GeForce 256 but at a reduced cost.

    GeForce 3 series

    In 2001 NVIDIA launched the GeForce3 (NV20) which introduced programmable vertex and pixel shaders. A version of GeForce 3 codenamed NV2A was developed for the Microsoft Xbox game console.

    GeForce 4 series

    In February 2002 the GeForce4 Ti (NV25) was launched as a refinement to the GeForce3. It included enhancements to anti-aliasing capabilities, an improved memory controller, a second vertex shader, and a manufacturing process size reduction to increase clock speeds. The GeForce4 MX was also introduced as a budget option based on the GeForce2, with the addition of some features from the GeForce4 Ti.

    GeForce FX series

    The GeForce FX (NV30) introduced a big change in architecture. It brought support for the new Shader Model 2 specification and carried the 5000 model number, as it was the fifth generation of the GeForce family. The series was also infamous for its heating and noisy fan issues.

    GeForce 6 series

    Launched in April 2004, the GeForce 6 (NV40) added Shader Model 3.0 and fixed the weak floating point shader performance of its predecessor. It additionally implemented high-dynamic-range imaging, SLI (Scalable Link Interface), and PureVideo capability (integrated partial hardware MPEG-2, VC-1, Windows Media Video, and H.264 decoding and fully accelerated video post-processing).

    GeForce 7 series

    The GeForce 7 series (G70/NV47) was introduced in June 2005 and was the last NVIDIA GPU series to support the AGP bus. It offered a wider pipeline and an increase in clock speed along with new transparency supersampling and transparency multisampling anti-aliasing modes (TSAA and TMAA). A version of the 7950 GT, called the RSX 'Reality Synthesizer', was used as the primary GPU on the Sony PlayStation 3.

    GeForce 8 series

    The first GeForce (G80) to fully support Direct3D 10, the 8th-gen GeForce series was launched in 2006. It was made using a 90nm process and built around the new Tesla microarchitecture. It was eventually refined and the die size was shrunk down to 65nm. The revised design codenamed G92 was implemented into the 8 series with the 8800GS, 8800GT, and 8800GTS-512, and was launched in 2007.

    GeForce 9 series

    Revisions for the GeForce 8 series were introduced after a short period in 2008 where the 9800GX2 used two G92 GPUs, in a dual PCB configuration with a single PCI-Express 16x slot. It also included two separate 256-bit memory busses, one for each GPU and a total of 1GB of memory on the card. Later the 9800GTX was launched with a single G92 GPU, 256-bit data bus, and 512 MB of GDDR3 memory.

    GeForce 100 series

    The following year NVIDIA launched the GeForce 100 series which were essentially rebranded versions of the GeForce 9 series available only for OEMs, although the GTS 150 was briefly available to consumers.

    GeForce 200

    The GeForce 200 series included the GT200 65nm-based graphics processor that had a total of 1.4 billion transistors and was introduced in 2008. It was also the year when NVIDIA changed its card-naming scheme by replacing the series number with the GTX or GTS suffix and then adding model numbers after that. The series features the new GT200 core on a 65nm die. The GeForce GTX 260 and the GTX 280 were the first products in the series while the GeForce 310 was released in November 2009 as a rebrand of GeForce 210.

    GeForce 300 series

    The 300 series cards were launched during the same year and were pretty much rebranded versions of the 200 series with support for DirectX 10.1 and based on the newer Fermi architecture. These were limited to OEMs only.

    GeForce 400 series

    The GeForce 400 series, codenamed GF100 was introduced in 2010 based on the Fermi architecture. They were the first NVIDIA GPUs to utilize 1GB or more of GDDR5 memory. The GTX 470 and GTX 480 were criticized for their high power use, high temperatures, and loud noise. At the same time, the GTX 480 was the fastest DirectX 11 card.

    GeForce 500 series

    To fix the issues, NVIDIA brought the 500 series with a new flagship (GTX 580) GPU based on an enhanced GF100 architecture (GF110). It offered higher performance, less power utilization, heat, and noise than the preceding GTX 480. Additionally, the GTX 590 was also introduced that packed two GF110 GPUs on a single card.

    GeForce 600 series

    In 2010, NVIDIA announced the Kepler microarchitecture, manufactured with the TSMC 28nm fabrication process. The company started supplying their top-end GK110 cores for use in Oak Ridge National Laboratory's Titan supercomputer, leading to a shortage of GK110 cores. Eventually, NVIDIA had to use the GK104 core, which was originally intended for the mid-range segment, to power their flagship, the GTX 680. It was followed by the dual-GK104 GTX 690 and the GTX 670.

    GeForce 700 series

    In May 2013, NVIDIA announced the 700 series based on the Kepler architecture, although it finally featured a GK110 chipset-based card at the top of the lineup. The GTX 780 was a cut-down version of the GTX Titan that achieved nearly the same performance for two-thirds of the price. A week after the release of the GTX 780, NVIDIA announced the GTX 770 to be a rebrand of the GTX 680. It was followed by the GTX 760 which was also based on the GK104 core and similar to the GTX 660 Ti.

    GeForce 800M series

    The GeForce 800M series included rebranded 700M series parts based on the Kepler architecture and some lower-end parts based on the newer Maxwell architecture.

    GeForce 900 series

    In March 2013, NVIDIA announced the new Maxwell microarchitecture. It was released in September 2014 on the GeForce 900 series and was the last series to support analog video output through DVI-I.

    GeForce 10 series

    In March 2014, NVIDIA announced that the successor to Maxwell would be the Pascal microarchitecture and was finally introduced on the GeForce 10 series in May 2016. It included 128 CUDA cores per streaming multiprocessor, GDDR5X memory, unified memory, and NVLink.

    GeForce 20 series

    In August 2018, NVIDIA announced Turing architecture as a successor to Pascal. The new microarchitecture was made to accelerate the real-time ray tracing support and AI Inferencing. It included a new Ray Tracing unit (RT Core) which dedicated processors to the ray tracing in hardware. It also supported the DXR extension in Microsoft DirectX 12. The company also introduced DLSS (Deep Learning Super Sampling), a new form of anti-aliasing that used AI to provide sharper imagery with less impact on performance.

    The first GPUs to utilize the architecture were primarily aimed at high-end professionals and were introduced under the Quadro series. Eventually, the GeForce RTX series with RTX 2080 Ti, 2080, and 2070 were announced in 2018 followed by the RTX 2060 in January 2019.

    In July 2019, NVIDIA announced the GeForce RTX Super line of cards, a refresh of the RTX 20 series which featured higher-spec versions of the RTX 2060, 2070, and 2080.

    GeForce 16 series

    In February 2019, NVIDIA announced the GeForce 16 series. Based on the same Turing architecture used in the GeForce 20 series, this series omitted the Tensor (AI) and RT (ray tracing) cores. This series continues to offer a more affordable graphics solution for gamers while still attaining a higher performance compared to respective cards of the previous GeForce generations. Similar to the RTX Super refresh, NVIDIA announced the GTX 1650 Super and 1660 Super cards, in October 2019.

    GeForce 30 series

    The latest and the most powerful graphics cards from NVIDIA, the new 30-series take over from the 20-series and were announced in 2020. It introduced a massive jump over the predecessor and an excellent price to performance ratio. However, getting your hands on one is a difficult task.

    Mobile GPUs

    NVIDIA produced a wide range of graphics cards for notebooks as far as the GeForce 2 series, under the GeForce Go branding. Most of the features present in the desktop counterparts were made available on the mobile version. With the introduction of the GeForce 8 series, the GeForce Go brand was discontinued and mobile GPUs were now a part of the main GeForce GPUs, with an M suffix. Once again NVIDIA brought some changes and dropped the M suffix in 2016 with the launch of the laptop GeForce 10 series in an attempt to unify the branding between their desktop and laptop GPU offerings. Currently, the RTX 20, GTX 16 and RTX 30 series of GPUs are available as both desktop and laptop variants. NVIDIA also has the GeForce MX range of mobile GPUs intended for lightweight notebooks with entry-level performance.

    Nomenclature

    Ever since the launch of the GeForce 100 series NVIDIA has been using the following naming scheme for its products:

    G, GT, No Prefix - Mostly user for entry-level category of graphics cards with the last two numbers ranging from 00 to 45. Example - GeForce GT 730, GeForce GT 1030

    GTS, GTX, RTX - Mid-range category of graphics cards with the last two numbers ranging from 50 to 65. Example - GeForce GTX 1060, GeForce RTX 2060

    GTX, RTX - High-end range of graphics cards with the last two numbers ranging from 70-95. Example - GeForce GTX 1080Ti, GeForce RTX 3090

    NVIDIA also uses the ‘Super’ or ‘Ti’ suffixes for its graphics cards to signify incremental updates.
    1
    I am having some issues with my GeForce GTX 1050Ti
    My motherboard sometimes doesn't detect the card and I am forced to use the Intel inbuild GPU
    I was having the same issue with my gt 730 gddr5 card. I found out that the system memory was causing the problem.
    1
    I replaced my bad memory stick and voila the issue was fixed!!!