Nvidia GeForce 500 series
The GeForce 500 series is a series of desktop - graphics chip company Nvidia and successor of the GeForce 400 series . All graphics processors of this series support the Shader model 5.0 (SM 5.0), DirectX 11 as well as OpenCL , CUDA and thus also PhysX .
description
Nvidia presented the first graphics card of the GeForce 500 series on November 9, 2010. This is the GeForce GTX 580 that used the GF110 graphics processor. This is still based on the Fermi architecture of the GeForce 400 series , albeit in a slightly modified version, which Nvidia's CEO Jen-Hsun Huang described as a "mid-life kicker". Nevertheless, from a technical point of view, the GF110 is “only” a new edition of the GF100 with a revised stepping. Full-fledged FP16 filtering has been added and Z-Cull efficiency has been improved. In order to reduce leakage currents and thus lower power consumption, fast-switching transistors have been replaced by slower models. When it was presented, the GeForce GTX 580 had a performance that was around 15% to 20% higher than the GeForce GTX 480 and replaced it as the fastest single GPU card on the market. Since AMD's competitor model , the Radeon HD 6970 , which was published later , only achieved roughly the same performance as the GeForce GTX 480/570, the GTX 580 as the fastest single GPU card has a monopoly on the market. After the GTX 480 was criticized in the trade press for its high volume, among other things, Nvidia installed a new dual-slot cooling system on the GTX 580, which uses a specially developed vapor chamber technology made of copper. This means that the GeForce GTX 580 produced significantly less noise. With the second big point of criticism of the GeForce GTX 480, the high power consumption, Nvidia only achieved marginal improvements. The GeForce GTX 580 is the first graphics card from Nvidia in the high-end sector since the GeForce GTX 285 , in which the GPU is used in the full expansion stage. With the GeForce 400 series, this was not yet possible due to problems with the 40 nm manufacturing process at TSMC and excessive heat generation. The GeForce GTX 570 followed on December 7, 2010, in which a shader cluster of the GF110 graphics processor was deactivated. It achieved almost exactly the performance of the GeForce GTX 480, but had better values in the areas of power consumption and noise, but the graphics memory was also reduced from 1536 MB to 1280 MB.
A special feature of the series is the GeForce GTX 590, which was presented on March 24, 2011. It is Nvidia's first official dual-GPU graphics card since the GeForce GTX 295, as there is no such derivative in the GeForce 400 series on the market had been brought. It is based on two GF110 GPUs in full configuration, which are operated with significantly reduced clock rates compared to the GeForce GTX 580. Nevertheless, Nvidia officially specifies the TDP as 365 watts, which is outside the PCI-SIG specifications. This had already been broken a few weeks earlier by the Radeon HD 6990, with which the GTX 590 competed for the performance crown, with both cards reaching new negative records in the area of power consumption.
On January 25, 2011, Nvidia introduced the GeForce GTX 560 Ti as the successor to the successful GeForce GTX 460 . The GF114 graphics processor, which is a refresh of the GF104, now served as the basis. As with the GF110, fast-switching transistors were partially replaced by slower models in order to avoid leakage currents and thus reduce power consumption. In contrast to the GF104, the GF114 is used in full on the GeForce GTX 560 Ti, whereas a shader cluster was switched off on the GTX 460. Together with higher clock rates, the GTX 560 achieved about 30% higher 3D performance than the GTX 460. This placed it in the market between the AMD competitors Radeon HD 6870 and 6950 . With the GeForce GTX 560 Ti, Nvidia used the abbreviation Ti (Titanium) for the first time since the GeForce 4 series from 2002. Nvidia presented another model based on the GF114 in the form of the GeForce GTX 560 (without the abbreviation Ti) on May 17, 2011. A shader cluster of the GPU has now been deactivated, which ultimately means a higher-clocked version of the GeForce GTX 460 (including the improved stepping of the GF114). The GTX 560 achieves roughly the same performance as the Radeon HD 6870. Nvidia did not provide an official reference design for the card, but left this to the board partners. For both GTX-560 models, Nvidia presented special variants for the OEM market, which, however, differed greatly from the retail version in terms of technology, as they are based on partially deactivated GF110 GPUs.
On March 15, 2011, Nvidia introduced the GeForce GTX 550 Ti based on the GF116 GPU. The card represents the successor to the GeForce GTS 450, whereby Nvidia changed the naming scheme and gave up the abbreviation GTS. Instead, Nvidia now also uses the GTX abbreviation here, which has been criticized in the specialist press because it suggests that it belongs to the performance and high-end sector, which does not exist. A special feature of the GTX 550 Ti is the memory configuration: Although a 192-bit memory interface is used, the Vram size is 1024 MB. Nvidia achieves this by connecting 256 MB twice to two storage controllers and 512 MB once to the third. Regardless of this, the GeForce GTX 550 Ti achieves roughly the same performance as the Radeon HD 5770 (later renamed Radeon HD 6770), which was introduced 18 months earlier, although the GTX 550 requires more power. For the OEM market, Nvidia brought out the GeForce GT 545, in which a shader cluster of the GF116 GPU was deactivated.
On April 12, 2011, Nvidia introduced the GeForce GT 520, which uses the GF119 graphics processor, with the VP5 video processor being used for the first time. A speed-reduced version followed at the end of September under the designation GeForce 510. The GeForce GT 530 was introduced for the OEM market on May 11, 2011, a new edition of the GeForce GT 430 based on the GF108 graphics processor.
For the GeForce 500 series, the naming scheme introduced with the GeForce 200 desktop series was used. All graphics chips are identified with a letter abbreviation to classify the performance sector and a three-digit number that generally begins with a "5" (for GeForce 500). The last two digits serve for further differentiation within the respective service sector. However, the letter abbreviations in this series have largely lost their meaning. From mainstream to high-end, everything is sold under the abbreviation "GTX" , which once stood for high-end and performance . In the low-end segment there is the "GT" abbreviation.
Data overview
Graphics processors
Graphics chip |
production | units | L2 cache |
API support | Video pro- cessor |
Bus interface stelle |
|||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
production process |
transis- interfere |
The - area |
ROP particle functions |
ROPs | Unified shaders | Texture units | DirectX | OpenGL | OpenCL | ||||||
Stream processors |
Shader - cluster |
TAUs | TMUs | ||||||||||||
GF108 | 40 nm | 0.58 billion | 114 mm² | 1 | 4th | 96 | 2 | 16 | 16 | k. A. | 11.0 | 4.4 | 1.1 | VP4 | PCIe 2.0 |
GF110 | ≈ 3 billion | 520 mm² | 6th | 48 | 512 | 16 | 64 | 64 | 768 KB | ||||||
GF114 | 1.95 billion | 332 mm² | 4th | 32 | 384 | 8th | 64 | 64 | 512 KB | ||||||
GF116 | 1.17 billion | 238 mm² | 3 | 24 | 192 | 4th | 32 | 32 | 384 KB | ||||||
GF119 | 0.29 billion | 79 mm² | 1 | 4th | 48 | 1 | 8th | 8th | k. A. | VP5 |
Model data
model | Official launch |
Graphics processor (GPU) | Graphics memory | Performance data | ||||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Type | Active units | Chip clock (in MHz) |
Size (in MB ) |
Clock rate (in MHz) |
Type | Storage interface |
Computing power (in GFlops ) |
Polygon throughput (in million triangles / s) |
Pixel fill rate (in GPixel / s) |
Texel fill rate (in GTexel / s) |
Memory bandwidth (in GB / s) |
|||||||
ROPs |
Shader - cluster |
ALUs |
Texture units |
GPU | Shader | SP (MAD) | DP (FMA) | |||||||||||
GeForce 510 | 29 Sep 2011 | GF119 | 4th | 1 | 48 | 8th | 523 | 1046 | 1024-2048 | 898 | DDR3 | 64 bit | 100.4 | 8.4 | 131 | 2.1 | 4.2 | 14.4 |
GeForce GT 520 | May 17, 2011 | GF119 | 4th | 1 | 48 | 8th | 810 | 1620 | 1024-2048 | 898 | DDR3 | 64 bit | 155.5 | 13 | 203 | 3.2 | 6.5 | 14.4 |
GeForce GT 520 | Apr 12, 2011 | GF119 | 4th | 1 | 48 | 8th | 810 | 1620 | 1024 | 900 | DDR3 | 64 bit | 155.5 | 13 | 203 | 3.2 | 6.5 | 14.4 |
GeForce GT 530 | May 17, 2011 | GF108 | 4th | 2 | 96 | 16 | 700 | 1400 | 1024-2048 | 898 | DDR3 | 128 bit | 268.8 | 22.4 | 350 | 2.8 | 11.2 | 28.7 |
GeForce GT 545 DRR3 | May 17, 2011 | GF116 | 24 | 3 | 144 | 24 | 720 | 1440 | 1536-3072 | 900 | DDR3 | 192 bits | 414.7 | 34.6 | 540 | 8.6 | 17.3 | 43.2 |
GeForce GT 545 GDDR5 | May 17, 2011 | GF116 | 16 | 3 | 144 | 24 | 870 | 1740 | 1024 | 1998 (999) | GDDR5 | 128 bit | 466.6 | 38.9 | 652 | 9.7 | 19.4 | 63.9 |
GeForce GTX 550 Ti | March 15, 2011 | GF116 | 24 | 4th | 192 | 32 | 900 | 1800 | 1024 | 2052 (1026) | GDDR5 | 192 bits | 691.2 | 57.6 | 900 | 14.4 | 28.8 | 98.5 |
GeForce GTX 555 | Jan. 18, 2012 | GF114 | 24 | 6th | 288 | 32 | 776 | 1553 | 1024 | 1914 (957) | GDDR5 | 192 bits | 894.5 | 74.5 | 1164 | 18.6 | 37.2 | 91.9 |
GeForce GTX 560 | Jun 16, 2011 | GF110 | 40 | 12 | 384 | 48 | 552 | 1104 | 1280-2560 | 1603 (802) | GDDR5 | 320 bits | 847.9 | 106 | 1656 | 17.7 | 26.5 | 128.2 |
GeForce GTX 560 | May 17, 2011 | GF114 | 32 | 7th | 336 | 56 | 810 | 1620 | 1024 | 2004 (1002) | GDDR5 | 256 bit | 1088.6 | 90.7 | 1418 | 22.7 | 45.4 | 128.3 |
GeForce GTX 560 Ti | May 30, 2011 | GF110 | 40 | 11 | 352 | 44 | 732 | 1464 | 1280-2560 | 1900 (950) | GDDR5 | 320 bits | 1030.7 | 128.8 | 2013 | 16.1 | 32.2 | 152 |
GeForce GTX 560 Ti | Jan. 25, 2011 | GF114 | 32 | 8th | 384 | 64 | 822 | 1645 | 1024 | 2004 (1002) | GDDR5 | 256 bit | 1263.4 | 105.3 | 1645 | 26.3 | 52.6 | 128.3 |
GeForce GTX 560 Ti 448 Cores | Nov 29, 2011 | GF110 | 40 | 14th | 448 | 56 | 732 | 1464 | 1280 | 1900 (950) | GDDR5 | 320 bits | 1311.7 | 164 | 2562 | 20.5 | 41.3 | 152 |
GeForce GTX 570 | Dec 7, 2010 | GF110 | 40 | 15th | 480 | 60 | 732 | 1464 | 1280 | 1900 (950) | GDDR5 | 320 bits | 1405.4 | 175.7 | 2745 | 22nd | 43.9 | 152 |
GeForce GTX 580 | Nov 9, 2010 | GF110 | 48 | 16 | 512 | 64 | 772 | 1544 | 1536-3072 | 2004 (1002) | GDDR5 | 384 bits | 1581.1 | 197.6 | 3088 | 24.7 | 49.4 | 192.4 |
GeForce GTX 590 | March 24, 2011 | 2 × GF110 | 2 × 48 | 2 × 16 | 2 × 512 | 2 × 64 | 607 | 1215 | 2 × 1536 | 1707 (854) | GDDR5 | 2 × 384 bits | 2 x 1244.2 | 2 x 155.5 | 2 × 2430 | 2 x 19.4 | 2 x 38.8 | 2 x 163.9 |
Power consumption data
model | Type | Consumption ( watt ) | additional power plug |
|||
---|---|---|---|---|---|---|
MGCP |
Readings | |||||
Idle | 3D load |
Maximum load |
||||
GeForce 510 (OEM) | GF119 | 25th | k. A. | k. A. | k. A. | no |
GeForce GT 520 (OEM) | GF119 | 29 | k. A. | k. A. | k. A. | no |
GeForce GT 520 | GF119 | 29 | 7th | 34 | 36 | no |
GeForce GT 530 (OEM) | GF108 | 50 | k. A. | k. A. | k. A. | no |
GeForce GT 545 DDR3 (OEM) | GF116 | 70 | k. A. | k. A. | k. A. | no |
GeForce GT 545 GDDR5 (OEM) | GF116 | 105 | k. A. | k. A. | k. A. | 1 × 6 pin |
GeForce GTX 550 Ti | GF116 | 116 | 14-15 | 116-120 | 142-163 | 1 × 6 pin |
GeForce GTX 555 (OEM) | GF114 | 150 | k. A. | k. A. | k. A. | 2 × 6-pin |
GeForce GTX 560 (OEM) | GF110 | 150 | k. A. | k. A. | k. A. | 2 × 6-pin |
GeForce GTX 560 | GF114 | 150 | k. A. | k. A. | k. A. | 2 × 6-pin |
GeForce GTX 560 Ti (OEM) | GF110 | 210 | k. A. | k. A. | k. A. | 2 × 6-pin |
GeForce GTX 560 Ti | GF114 | 170 | 16 | 152-153 | 193-208 | 2 × 6-pin |
GeForce GTX 560 Ti 448 Cores | GF110 | 210 | 26th | 214 | 258 | 2 × 6-pin |
GeForce GTX 570 | GF110 | 219 | 24-28 | 194-199 | 247-287 | 2 × 6-pin |
GeForce GTX 580 | GF110 | 244 | 31-33 | 221-247 | 312-318 | 1 × 6-pin 1 × 8-pin |
GeForce GTX 590 | 2 × GF110 | 365 | 52-54 | 343-358 | 428-444 | 2 × 8 pin |
Remarks
- ↑ The date indicated is the date of the public presentation, not the date of availability of the models.
- ↑ The specified performance values for the computing power via the stream processors, the pixel and texel filling rate, as well as the memory bandwidth are theoretical maximum values (with standard clock rate) that cannot be directly compared with the performance values of other architectures. The overall performance of a graphics card depends, among other things, on how well the available resources can be used or fully utilized. There are also other factors that are not listed here that affect performance.
- ↑ a b The specified clock rates are the reference data recommended or specified by Nvidia; the I / O clock is specified for the memory clock. However, the exact clock rate can deviate by a few megahertz due to different clock generators, and the final definition of the clock rates is in the hands of the respective graphics card manufacturer. It is therefore entirely possible that there are or will be graphics card models that have different clock rates.
- ↑ a b c d e f g h OEM product. Card is not available in the retail market.
- ↑ The MGCP value specified by Nvidia does not necessarily correspond to the maximum power consumption. This value is also not necessarily comparable with the TDP value of the competitor AMD.
- ↑ The measured values listed in the table relate to the pure power consumption of graphics cards that correspond to the Nvidia reference design. A special measuring device is required to measure these values; Depending on the measurement technology used and the given measurement conditions, including the program used to generate the 3D load, the values can fluctuate between different devices. Therefore, measured value ranges are given here, each representing the lowest, typical and highest measured values from different sources.
- ↑ The value given under 3D load corresponds to the typical game usage of the card. However, this is different depending on the 3D application. As a rule, a modern 3D application is used to determine the value, which, however, limits the comparability over longer periods of time.
- ↑ The maximum load is usually determined with demanding benchmark programs, the loads of which are significantly higher than those of "normal" 3D applications.
Individual evidence
- ^ Nvidia: After Fermi follow Kepler and Maxwell. PC Games Hardware, September 22, 2010, accessed November 9, 2010 .
- ↑ a b Test: Nvidia GeForce GTX 580 - Technical Data. ComputerBase, November 9, 2010, accessed November 14, 2010 .
- ↑ a b test: Nvidia GeForce GTX 570 - performance rating. ComputerBase, December 7, 2010, accessed December 21, 2010 .
- ↑ Test: AMD Radeon HD 6970 and HD 6950 - performance rating. ComputerBase, December 15, 2010, accessed December 21, 2010 .
- ↑ Test: Nvidia GeForce GTX 580 - volume. ComputerBase, November 9, 2010, accessed November 14, 2010 .
- ↑ Review: Nvidia GeForce GTX 580 - power consumption. ComputerBase, November 9, 2010, accessed November 14, 2010 .
- ↑ Nvidia GeForce GTX 570 for 350 euros in the test: graphics quality with performance above GTX 480 level - impressions, loudness and power consumption. PC Games Hardware, December 7, 2010, accessed December 21, 2010 .
- ↑ Review: Nvidia GeForce GTX 590 - power consumption. ComputerBase, March 24, 2011, accessed May 13, 2011 .
- ↑ Review: Nvidia GeForce GTX 560 Ti (SLI) - Ratings. ComputerBase, January 25, 2011, accessed February 7, 2011 .
- ↑ Review: Nvidia GeForce GTX 560 - performance. ComputerBase, May 17, 2011, accessed May 8, 2011 .
- ↑ Test: 3 × Nvidia GeForce GTX 550 Ti (SLI) ratings. ComputerBase, March 15, 2011, accessed May 18, 2011 .
- ↑ GeForce GTX 550 Ti in the test: loudness and power consumption. PC Games Hardware, March 15, 2011, accessed May 8, 2011 .
- ↑ NVIDIA GeForce 510 (OEM). Nvidia Corporation, accessed October 6, 2011 .
- ↑ NVIDIA GeForce GT 520 (OEM). Nvidia Corporation, accessed May 30, 2011 .
- ↑ NVIDIA GeForce GT 520. Nvidia Corporation, accessed April 12, 2011 .
- ↑ NVIDIA GeForce GT 530 (OEM). Nvidia Corporation, accessed May 30, 2011 .
- ↑ NVIDIA GeForce GT 545 DDR3 (OEM). Nvidia Corporation, accessed May 30, 2011 .
- ↑ NVIDIA GeForce GT 545 GDDR5 (OEM). Nvidia Corporation, accessed May 30, 2011 .
- ↑ NVIDIA GeForce GTX 550 Ti. Nvidia Corporation, accessed March 15, 2011 .
- ↑ NVIDIA GeForce GTX 555 (OEM). Nvidia Corporation, accessed January 20, 2012 .
- ↑ NVIDIA GeForce GTX 560 (OEM). Nvidia Corporation, accessed June 18, 2011 .
- ↑ NVIDIA GeForce GTX 560. Nvidia Corporation, accessed May 17, 2011 .
- ↑ NVIDIA GeForce GTX 560 Ti (OEM). Nvidia Corporation, accessed June 1, 2011 .
- ↑ a b NVIDIA GeForce GTX 560 Ti. Nvidia Corporation, accessed January 25, 2011 .
- ↑ NVIDIA GeForce GTX 570. Nvidia Corporation, accessed December 7, 2010 .
- ↑ NVIDIA GeForce GTX 580. Nvidia Corporation, accessed November 9, 2010 .
- ↑ EVGA GeForce GTX 580 3072MB. Retrieved January 14, 2020 .
- ↑ NVIDIA GeForce GTX 590. Nvidia Corporation, accessed March 24, 2011 .
- ↑ a b c d e f Zotac GeForce GT 520 1 GByte DDR3 in the test - power consumption: idle & load. HardTecs4U, November 29, 2011, accessed May 1, 2011 .
- ↑ a b c GeForce GTX 550 Ti in the test: The better Radeon HD 5770? - Impressions, loudness and power consumption. PC Games Hardware, March 15, 2011, accessed March 15, 2011 .
- ↑ a b c d e f g h i j k l Gigabyte GeForce GTX 560 Ti 448 in the test - power consumption: idle & load. HardTecs4U, November 29, 2011, accessed November 29, 2011 .
- ↑ a b c d e f GeForce GTX 590 in the test: Does the double Fermi triumph over the Radeon HD 6990? - Impressions, loudness and power consumption. PC Games Hardware, March 24, 2011, accessed March 24, 2011 .
- ↑ a b c d e f Nvidia GeForce GTX 570 for 350 euros in the test: graphics quality with performance above GTX 480 level - impressions, loudness and power consumption. PC Games Hardware, December 7, 2010, accessed December 7, 2010 .
- ↑ a b c NVIDIA GeForce GTX 590 in the test - reaching for the crown: power consumption. HardTecs4U, March 24, 2011, accessed October 12, 2012 .