It is possible to use Panda Dataframes to scrape directly an HTML table from a URL. This can be incredibly useful and a quick way to extract useful data, when you are in a hurry.
To see how valuable it is, let’s use a concrete example. Imagine that we want to scrape some information about all the GPUs that Nvidia launched recently.
Wikipedia here comes to the rescue as we have fairly accurate information available in this url:
https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units
We want to extract the table below:
Let’s first import the pandas library
import pandasChecking the Python docs for pandas.read_html
Before we try reading an HTML table, let’s check the documentation. Here is a quick way to do that directly from Python.
help(pandas.read_html)Help on function read_html in module pandas.io.html:
read_html(io:Union[str, pathlib.Path, IO[~AnyStr]], match:Union[str, Pattern[~AnyStr]]='.+', flavor:Union[str, NoneType]=None, header:Union[int, Sequence[int], NoneType]=None, index_col:Union[int, Sequence[int], NoneType]=None, skiprows:Union[int, Sequence[int], slice, NoneType]=None, attrs:Union[Dict[str, str], NoneType]=None, parse_dates:bool=False, thousands:Union[str, NoneType]=',', encoding:Union[str, NoneType]=None, decimal:str='.', converters:Union[Dict, NoneType]=None, na_values=None, keep_default_na:bool=True, displayed_only:bool=True) -> List[pandas.core.frame.DataFrame]
Read HTML tables into a ``list`` of ``DataFrame`` objects.
Parameters
----------
io : str, path object or file-like object
A URL, a file-like object, or a raw string containing HTML. Note that
lxml only accepts the http, ftp and file url protocols. If you have a
URL that starts with ``'https'`` you might try removing the ``'s'``.
match : str or compiled regular expression, optional
The set of tables containing text matching this regex or string will be
returned. Unless the HTML is extremely simple you will probably need to
pass a non-empty string here. Defaults to '.+' (match any non-empty
string). The default value will return all tables contained on a page.
This value is converted to a regular expression so that there is
consistent behavior between Beautiful Soup and lxml.
flavor : str, optional
The parsing engine to use. 'bs4' and 'html5lib' are synonymous with
each other, they are both there for backwards compatibility. The
default of ``None`` tries to use ``lxml`` to parse and if that fails it
falls back on ``bs4`` + ``html5lib``.
header : int or list-like, optional
The row (or list of rows for a :class:`~pandas.MultiIndex`) to use to
make the columns headers.
index_col : int or list-like, optional
The column (or list of columns) to use to create the index.
skiprows : int, list-like or slice, optional
Number of rows to skip after parsing the column integer. 0-based. If a
sequence of integers or a slice is given, will skip the rows indexed by
that sequence. Note that a single element sequence means 'skip the nth
row' whereas an integer means 'skip n rows'.
attrs : dict, optional
This is a dictionary of attributes that you can pass to use to identify
the table in the HTML. These are not checked for validity before being
passed to lxml or Beautiful Soup. However, these attributes must be
valid HTML table attributes to work correctly. For example, ::
attrs = {'id': 'table'}
is a valid attribute dictionary because the 'id' HTML tag attribute is
a valid HTML attribute for *any* HTML tag as per `this document
<https://html.spec.whatwg.org/multipage/dom.html#global-attributes>`__. ::
attrs = {'asdf': 'table'}
is *not* a valid attribute dictionary because 'asdf' is not a valid
HTML attribute even if it is a valid XML attribute. Valid HTML 4.01
table attributes can be found `here
<http://www.w3.org/TR/REC-html40/struct/tables.html#h-11.2>`__. A
working draft of the HTML 5 spec can be found `here
<https://html.spec.whatwg.org/multipage/tables.html>`__. It contains the
latest information on table attributes for the modern web.
parse_dates : bool, optional
See :func:`~read_csv` for more details.
thousands : str, optional
Separator to use to parse thousands. Defaults to ``','``.
encoding : str, optional
The encoding used to decode the web page. Defaults to ``None``.``None``
preserves the previous encoding behavior, which depends on the
underlying parser library (e.g., the parser library will try to use
the encoding provided by the document).
decimal : str, default '.'
Character to recognize as decimal point (e.g. use ',' for European
data).
converters : dict, default None
Dict of functions for converting values in certain columns. Keys can
either be integers or column labels, values are functions that take one
input argument, the cell (not column) content, and return the
transformed content.
na_values : iterable, default None
Custom NA values.
keep_default_na : bool, default True
If na_values are specified and keep_default_na is False the default NaN
values are overridden, otherwise they're appended to.
displayed_only : bool, default True
Whether elements with "display: none" should be parsed.
Returns
-------
dfs
A list of DataFrames.
See Also
--------
read_csv : Read a comma-separated values (csv) file into DataFrame.
Notes
-----
Before using this function you should read the :ref:`gotchas about the
HTML parsing libraries <io.html.gotchas>`.
Expect to do some cleanup after you call this function. For example, you
might need to manually assign column names if the column names are
converted to NaN when you pass the `header=0` argument. We try to assume as
little as possible about the structure of the table and push the
idiosyncrasies of the HTML contained in the table to the user.
This function searches for ``<table>`` elements and only for ``<tr>``
and ``<th>`` rows and ``<td>`` elements within each ``<tr>`` or ``<th>``
element in the table. ``<td>`` stands for "table data". This function
attempts to properly handle ``colspan`` and ``rowspan`` attributes.
If the function has a ``<thead>`` argument, it is used to construct
the header, otherwise the function attempts to find the header within
the body (by putting rows with only ``<th>`` elements into the header).
Similar to :func:`~read_csv` the `header` argument is applied
**after** `skiprows` is applied.
This function will *always* return a list of :class:`DataFrame` *or*
it will fail, e.g., it will *not* return an empty list.
Examples
--------
See the :ref:`read_html documentation in the IO section of the docs
<io.read_html>` for some examples of reading in HTML tables.The important parameters are:
- str – a url to the website with the table you want to extract, or just the HTML of the page as a string.
- match – a regular expression that identifies the table that you want to extract
- flavor – this picks the engine doing the scraping. bs4 stands for BeautifulSoup.
Let’s try to import a table with pandas.read_html:
tables =pandas.read_html("https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units")
display(tables)[ Model Launch ... API support
Model Launch ... Direct3D OpenGL
0 STG-2000 May 22, 1995 ... NaN NaN
1 Riva 128 August 25, 1997 ... 5.0 1.0
2 Riva 128ZX February 23, 1998 ... 5.0 1.0
3 Riva TNT June 15, 1998 ... 6.0 1.2
4 Vanta March 22, 1999 ... 6.0 1.2
5 Vanta LT March 2000 ... 6.0 1.2
6 Riva TNT2 M64 October 1999 ... 6.0 1.2
7 Riva TNT2 March 15, 1999 ... 6.0 1.2
8 Riva TNT2 Pro October 12, 1999 ... 6.0 1.2
9 Riva TNT2 Ultra March 15, 1999 ... 6.0 1.2
10 Model Launch ... Direct3D OpenGL
11 Model Launch ... API support API support
[12 rows x 18 columns],
Model Launch ... Memory Processing power (GFLOPS)
Model Launch ... Bus width (bit) Single precision
0 GeForce 256 SDR October 11, 1999 ... 128 50[4]
1 GeForce 256 DDR December 13, 1999 ... 128 50[4]
[2 rows x 17 columns],
Model ... Memory
Model ... Bus width (bit)
0 GeForce2 MX IGP + nForce 220/420 ... 64128
1 GeForce2 MX200 ... 64
2 GeForce2 MX ... 128
3 GeForce2 MX400 ... 128 (SDR)64 (DDR)
4 GeForce2 GTS ... 128
5 GeForce2 Pro ... 128
6 GeForce2 Ti ... 128
7 GeForce2 Ultra ... 128
8 Model ... Bus width (bit)
9 Model ... Memory
[10 rows x 16 columns],
Model Launch ... Memory
Model Launch ... Bus type Bus width (bit)
0 GeForce3 Ti200 October 1, 2001 ... DDR 128
1 GeForce3 February 27, 2001 ... DDR 128
2 GeForce3 Ti500 October 1, 2001 ... DDR 128
[3 rows x 16 columns],
Model ... Supported API version
Model ... OpenGL
0 GeForce4 MX IGP + nForce2 ... 1.2
1 GeForce4 MX420 ... 1.2
2 GeForce4 MX440 SE ... 1.2
3 GeForce MX4000 ... 1.2
4 GeForce PCX4300 ... 1.2
5 GeForce4 MX440 ... 1.2
6 GeForce4 MX440 8x ... 1.2
7 GeForce4 MX460 ... 1.2
8 GeForce4 Ti4200 ... 1.3
9 GeForce4 Ti4200 8x ... 1.3
10 GeForce4 Ti4400 ... 1.3
11 GeForce4 Ti4400 8x (Ti4800SE2) ... 1.3
12 GeForce4 Ti4600 ... 1.3
13 GeForce4 Ti4600 8x (Ti48003) ... 1.3
14 Model ... Supported API version
15 Model ... OpenGL
[16 rows x 18 columns],
Model Features
Model nFiniteFX II Engine Video Processing Engine (VPE)
0 GeForce4 MX420 No Yes
1 GeForce4 MX440 SE No Yes
2 GeForce4 MX4000 No Yes
3 GeForce4 PCX4300 No Yes
4 GeForce4 MX440 No Yes
5 GeForce4 MX440 8X No Yes
6 GeForce4 MX460 No Yes
7 GeForce4 Ti4200 Yes No
8 GeForce4 Ti4200 8x Yes No
9 GeForce4 Ti4400 Yes No
10 GeForce4 Ti4400 8x Yes No
11 GeForce4 Ti4600 Yes No
12 GeForce4 Ti4600 8x Yes No,
Model ... Memory
Model ... Bus width (bit)
0 GeForce FX 5100 ... 64
1 GeForce FX 5200 LE ... 64128
2 GeForce FX 5200 ... 64128
3 GeForce FX 5200 Ultra ... 128
4 GeForce PCX 5300 ... 64
5 GeForce FX 5500 ... 64 128
6 GeForce FX 5600 XT ... 64 128
7 GeForce FX 5600 ... 128
8 GeForce FX 5600 Ultra ... 128
9 GeForce FX 5600 Ultra Rev.2 ... 128
10 GeForce FX 5700 VE ... 64128
11 GeForce FX 5700 LE ... 64128
12 GeForce FX 5700 ... 128
13 GeForce PCX 5750 ... 128
14 GeForce FX 5700 Ultra ... 128
15 GeForce FX 5700 Ultra GDDR3 ... 128
16 GeForce FX 5800 ... 128
17 GeForce FX 5800 Ultra ... 128
18 GeForce FX 5900 ZT ... 256
19 GeForce FX 5900 XT ... 256
20 GeForce FX 5900 ... 256
21 GeForce FX 5900 Ultra ... 256
22 GeForce PCX 5900 ... 256
23 GeForce FX 5950 Ultra ... 256
24 GeForce PCX 5950 ... 256
25 Model ... Memory
26 Model ... Bus width (bit)
[27 rows x 16 columns],
Model ... Memory
Model ... Bus width (bit)
0 GeForce 6100 + nForce 410 ... 64128
1 GeForce 6150 SE + nForce 430 ... 64128
2 GeForce 6150 LE + nForce 430 ... 64128
3 GeForce 6150 + nForce 430 ... 64128
4 GeForce 6200 LE ... 64
5 GeForce 6200A ... 64[8]
6 GeForce 6200 ... 128
7 GeForce 6200 TurboCache ... 64
8 GeForce 6500 ... 64
9 GeForce 6600 LE ... 128
10 GeForce 6600 ... 128
11 GeForce 6600 GT ... 128
12 GeForce 6800 LE ... 256
13 GeForce 6800 XT ... 64[11] 128[12] 256
14 GeForce 6800 ... 256
15 GeForce 6800 GTO ... 256
16 GeForce 6800 GS ... 256
17 GeForce 6800 GT ... 256
18 GeForce 6800 Ultra ... 256
19 GeForce 6800 Ultra Extreme Edition ... 256
20 Model ... Memory
21 Model ... Bus width (bit)
[22 rows x 16 columns],
Model Features ...
Model OpenEXR HDR ... Unnamed: 5_level_1 Unnamed: 6_level_1
0 GeForce 6100 No ... NaN NaN
1 GeForce 6150 SE No ... NaN NaN
2 GeForce 6150 No ... NaN NaN
3 GeForce 6150 LE No ... NaN NaN
4 GeForce 6200 No ... NaN NaN
5 GeForce 6500 No ... NaN NaN
6 GeForce 6600 LE Yes ... NaN NaN
7 GeForce 6600 Yes ... NaN NaN
8 GeForce 6600 DDR2 Yes ... NaN NaN
9 GeForce 6600 GT Yes ... NaN NaN
10 GeForce 6800 LE Yes ... NaN NaN
11 GeForce 6800 XT Yes ... NaN NaN
12 GeForce 6800 Yes ... NaN NaN
13 GeForce 6800 GTO Yes ... NaN NaN
14 GeForce 6800 GS Yes ... NaN NaN
15 GeForce 6800 GT Yes ... NaN NaN
16 GeForce 6800 Ultra Yes ... NaN NaN
[17 rows x 7 columns],
Model ... Memory
Model ... Bus width (bit)
0 GeForce 7025 + nForce 630a ... 64 128
1 GeForce 7050PV + nForce 630a ... 64 128
2 GeForce 7050 + nForce 610i/630i ... 64
3 GeForce 7100 + nForce 630i ... 64
4 GeForce 7150 + nForce 630i ... 64
5 GeForce 7100 GS ... 32 64
6 GeForce 7200 GS ... 32 64
7 GeForce 7300 SE ... 32 64
8 GeForce 7300 LE ... 32 64
9 GeForce 7300 GS ... 64
10 GeForce 7300 GT ... 128
11 GeForce 7500 LE ... 3264
12 GeForce 7600 GS ... 128
13 GeForce 7600 GT ... 128
14 GeForce 7600 GT 80 nm ... 128
15 GeForce 7650 GS ... 128
16 GeForce 7800 GS ... 256
17 GeForce 7800 GT ... 256
18 GeForce 7800 GTX ... 256
19 GeForce 7900 GS ... 256
20 GeForce 7900 GT ... 256
21 GeForce 7900 GTO ... 256
22 GeForce 7900 GTX ... 256
23 GeForce 7900 GX2 ... 256
24 GeForce 7950 GT ... 256
25 GeForce 7950 GX2 ... 256
26 Model ... Memory
27 Model ... Bus width (bit)
[28 rows x 16 columns],
Model ... Features
Model ... Unnamed: 6_level_1
0 GeForce 7100 GS ... NaN
1 GeForce 7200 GS ... NaN
2 GeForce 7300 SE ... NaN
3 GeForce 7300 LE ... NaN
4 GeForce 7300 GS ... NaN
5 GeForce 7300 GT ... NaN
6 GeForce 7600 GS ... NaN
7 GeForce 7600 GT ... NaN
8 GeForce 7600 GT (80 nm) ... NaN
9 GeForce 7650 GS (80 nm) ... NaN
10 GeForce 7800 GS ... NaN
11 GeForce 7800 GT ... NaN
12 GeForce 7800 GTX ... NaN
13 GeForce 7800 GTX 512 ... NaN
14 GeForce 7900 GS ... NaN
15 GeForce 7900 GT ... NaN
16 GeForce 7900 GTO ... NaN
17 GeForce 7900 GTX ... NaN
18 GeForce 7900 GX2 (GTX Duo) ... NaN
19 GeForce 7950 GT ... NaN
20 GeForce 7950 GX2 ... NaN
[21 rows x 7 columns],
Model ... Comments
Model ... Comments
0 GeForce 8100 mGPU[15] ... The block of decoding of HD-video PureVideo HD...
1 GeForce 8200 mGPU[15] ... PureVideo 3 with VP3
2 GeForce 8300 mGPU[15] ... PureVideo 3 with VP3
3 GeForce 8300 GS[16] ... OEM only
4 GeForce 8400 GS ... NaN
5 GeForce 8400 GS rev.2 ... NaN
6 GeForce 8400 GS rev.3 ... NaN
7 GeForce 8500 GT ... NaN
8 GeForce 8600 GS ... OEM only
9 GeForce 8600 GT ... NaN
10 GeForce 8600 GTS ... NaN
11 GeForce 8800 GS ... NaN
12 GeForce 8800 GTS (G80) ... NaN
13 GeForce 8800 GTS 112 (G80) ... only XFX, EVGA and BFG models, very short-live...
14 GeForce 8800 GT ... NaN
15 GeForce 8800 GTS (G92) ... NaN
16 GeForce 8800 GTX ... NaN
17 GeForce 8800 Ultra ... NaN
18 Model ... Comments
19 Model ... Comments
[20 rows x 22 columns],
Model ... Features
Model ... Unnamed: 9_level_1
0 GeForce 8300 GS (G86) ... NaN
1 GeForce 8400 GS Rev. 2 (G98) ... NaN
2 GeForce 8400 GS Rev. 3 (GT218) ... NaN
3 GeForce 8500 GT ... NaN
4 GeForce 8600 GT ... NaN
5 GeForce 8600 GTS ... NaN
6 GeForce 8800 GS (G92) ... NaN
7 GeForce 8800 GTS (G80) ... NaN
8 GeForce 8800 GTS Rev. 2 (G80) ... NaN
9 GeForce 8800 GT (G92) ... NaN
10 GeForce 8800 GTS (G92) ... NaN
11 GeForce 8800 GTX ... NaN
12 GeForce 8800 Ultra ... NaN
[13 rows x 10 columns],
Model ... Comments
Model ... Comments
0 GeForce 9300 mGPU ... based on 8400 GS
1 GeForce 9400 mGPU ... based on 8400 GS
2 GeForce 9300 GE[18] ... NaN
3 GeForce 9300 GS[18] ... NaN
4 GeForce 9400 GT ... NaN
5 GeForce 9500 GT ... NaN
6 GeForce 9600 GS ... OEM
7 GeForce 9600 GSO ... NaN
8 GeForce 9600 GSO 512 ... NaN
9 GeForce 9600 GT Green Edition ... Core Voltage = 1.00v
10 GeForce 9600 GT ... NaN
11 GeForce 9800 GT Green Edition ... Core Voltage = 1.00v
12 GeForce 9800 GT ... NaN
13 GeForce 9800 GTX ... NaN
14 GeForce 9800 GTX+ ... NaN
15 GeForce 9800 GX2 ... NaN
16 Model ... Comments
[17 rows x 22 columns],
Model ... Features
Model ... Unnamed: 9_level_1
0 GeForce 9300 GE (G98) ... NaN
1 GeForce 9300 GS (G98) ... NaN
2 GeForce 9400 GT ... NaN
3 GeForce 9500 GT ... NaN
4 GeForce 9600 GSO ... NaN
5 GeForce 9600 GT ... NaN
6 GeForce 9800 GT ... NaN
7 GeForce 9800 GTX ... NaN
8 GeForce 9800 GTX+ ... NaN
9 GeForce 9800 GX2 ... NaN
[10 rows x 10 columns],
Model Launch ... TDP (Watts) Comments
Model Launch ... TDP (Watts) Comments
0 GeForce G 100 March 10, 2009 ... 35 OEM products
1 GeForce GT 120 March 10, 2009 ... 50 OEM products
2 GeForce GT 130 March 10, 2009 ... 75 OEM products
3 GeForce GT 140 March 10, 2009 ... 105 OEM products
4 GeForce GTS 150 March 10, 2009 ... 141 OEM products
[5 rows x 22 columns],
Model ... Release Price (USD)
Model ... Release Price (USD)
0 GeForce 205 ... NaN
1 GeForce 210 ... NaN
2 GeForce GT 220 ... NaN
3 GeForce GT 230 ... NaN
4 GeForce GT 230 ... NaN
5 GeForce GT 240 ... NaN
6 GeForce GTS 240 ... NaN
7 GeForce GTS 250 ... NaN
8 GeForce GTS 250 ... $150($130 512MB)
9 GeForce GTX 260 ... $400 (dropped to $270 after 3 months[22])
10 GeForce GTX 260 ... $300
11 GeForce GTX 275 ... $250
12 GeForce GTX 280 ... $650 (dropped to $430 after 3 months[23])
13 GeForce GTX 285 ... $400
14 GeForce GTX 295 ... $500
15 Model ... Release Price (USD)
16 Model ... Release Price (USD)
[17 rows x 23 columns],
Model ... Features
Model ... Unnamed: 8_level_1
0 GeForce 210 ... NaN
1 GeForce GT 220 ... NaN
2 GeForce GT 240 ... NaN
3 GeForce GTS 250 ... NaN
4 GeForce GTX 260 ... NaN
5 GeForce GTX 260 Core 216 ... NaN
6 GeForce GTX 260 Core 216 (55 nm) ... NaN
7 GeForce GTX 275 ... NaN
8 GeForce GTX 280 ... NaN
9 GeForce GTX 285 ... NaN
10 GeForce GTX 295 ... NaN
[11 rows x 9 columns],
Model ... Comments
Model ... Comments
0 GeForce 310 ... OEM Card, similar to Geforce 210
1 GeForce 315 ... OEM Card, similar to Geforce GT220
2 GeForce GT 320 ... OEM Card
3 GeForce GT 330[24] ... Specifications vary depending on OEM, similar ...
4 GeForce GT 330[24] ... Specifications vary depending on OEM, similar ...
5 GeForce GT 330[24] ... Specifications vary depending on OEM, similar ...
6 GeForce GT 340 ... OEM Card, similar to GT240
[7 rows x 20 columns],
Model ... Release Price (USD)
Model ... Release Price (USD)
0 GeForce 4057 ... OEM
1 GeForce GT 420 ... OEM
2 GeForce GT 430 ... OEM
3 GeForce GT 430 ... $79
4 GeForce GT 430 ... $79
5 GeForce GT 440 ... $100
6 GeForce GT 440 ... OEM
7 GeForce GTS 450 ... OEM
8 GeForce GTS 450 ... $129
9 GeForce GTX 460 SE ... $160
10 GeForce GTX 460 ... OEM
11 GeForce GTX 460 ... $199
12 GeForce GTX 460 ... $229
13 GeForce GTX 460 ... $199
14 GeForce GTX 465 ... $279
15 GeForce GTX 470 ... $349
16 GeForce GTX 480 ... $499
17 Model ... Release Price (USD)
18 Model ... Release Price (USD)
[19 rows x 25 columns],
Model ... Release Price (USD)
Model ... Release Price (USD)
0 GeForce 510 ... OEM
1 GeForce GT 520 ... $59
2 GeForce GT 530 ... OEM
3 GeForce GT 545 ... $149
4 GeForce GT 545 ... OEM
5 GeForce GTX 550 Ti ... $149
6 GeForce GTX 555 ... OEM
7 GeForce GTX 560 SE ... OEM
8 GeForce GTX 560 ... $199
9 GeForce GTX 560 Ti ... $249
10 GeForce GTX 560 Ti ... OEM
11 GeForce GTX 560 Ti 448 Cores ... $289
12 GeForce GTX 570 ... $349
13 GeForce GTX 580 ... $499
14 GeForce GTX 590 ... $699
15 Model ... Release Price (USD)
16 Model ... Release Price (USD)
[17 rows x 26 columns],
Model ... Release Price (USD)
Model ... Release Price (USD)
0 GeForce 6052 ... OEM
1 GeForce GT 6103 ... Retail
2 GeForce GT 6204 ... OEM
3 GeForce GT 6204 ... Retail
4 GeForce GT 625 ... OEM
5 GeForce GT 630 ... OEM
6 GeForce GT 630 ... Retail
7 GeForce GT 630 ... Retail
8 GeForce GT 630 ... NaN
9 GeForce GT 635 ... OEM
10 GeForce GT 6408 ... OEM
11 GeForce GT 6408 ... OEM
12 GeForce GT 6408 ... $100
13 GeForce GT 6408 ... OEM
14 GeForce GT 6408 ... NaN
15 GeForce GT 6459 ... OEM
16 GeForce GTX 645 ... OEM
17 GeForce GTX 650 ... $110
18 GeForce GTX 650 Ti ... $150 (130)
19 GeForce GTX 650 Ti Boost ... $170 (150)
20 GeForce GTX 660 ... $230 (180)
21 GeForce GTX 660 ... OEM
22 GeForce GTX 660 Ti ... $300
23 GeForce GTX 670 ... $400
24 GeForce GTX 680 ... $500
25 GeForce GTX 690 ... $1000
26 Model ... Release Price (USD)
27 Model ... Release Price (USD)
[28 rows x 28 columns],
Model ... Release Price (USD)
Model ... Release Price (USD)
0 GeForce GT 705[42]4 ... OEM
1 GeForce GT 710[43] ... OEM
2 GeForce GT 710[43] ... $35–45
3 GeForce GT 720[44] ... $49–59
4 GeForce GT 730[45]5,6 ... $69–79
5 GeForce GT 730[45]5,6 ... $69–79
6 GeForce GT 730[45]5,6 ... $69–79
7 GeForce GT 7407 ... $89–99
8 GeForce GT 7407 ... $89–99
9 GeForce GTX 745 ... OEM
10 GeForce GTX 750 ... $119
11 GeForce GTX 750 Ti ... $149
12 GeForce GTX 760 192-bit ... OEM
13 GeForce GTX 760 ... $249 ($219)
14 GeForce GTX 760 Ti8 ... OEM
15 GeForce GTX 770 ... $399 ($329)
16 GeForce GTX 780 ... $649 ($499)
17 GeForce GTX 780 Ti[50][51][52] ... $699
18 GeForce GTX TITAN[53][54][55] ... $999
19 GeForce GTX TITAN Black ... $999
20 GeForce GTX TITAN Z ... $2999
21 Model ... Release Price (USD)
22 Model ... Release Price (USD)
[23 rows x 27 columns],
Model ... Release price (USD)
Model ... MSRP
0 GeForce GT 945A[61][62][63] ... OEM
1 GeForce GTX 950[64] ... $159
2 GeForce GTX 950 (OEM)[66] ... OEM
3 GeForce GTX 960[67] ... $199
4 GeForce GTX 960 (OEM)[69] ... OEM
5 GeForce GTX 970[70] ... $329
6 GeForce GTX 980[72] ... $549
7 GeForce GTX 980 Ti[73] ... $649
8 GeForce GTX TITAN X[74] ... $999
[9 rows x 23 columns],
Model ... Release price (USD)
Model ... Founders Edition
0 GeForce GT 1010[79] ... NaN
1 GeForce GT 1030[80][81] ... NaN
2 GeForce GT 1030[80][81] ... NaN
3 GeForce GTX 1050[84] ... NaN
4 GeForce GTX 1050[84] ... NaN
5 GeForce GTX 1050 Ti[84] ... NaN
6 GeForce GTX 1060[85][86][87][88] ... NaN
7 GeForce GTX 1060[85][86][87][88] ... NaN
8 GeForce GTX 1060[85][86][87][88] ... NaN
9 GeForce GTX 1060[85][86][87][88] ... NaN
10 GeForce GTX 1060[85][86][87][88] ... NaN
11 GeForce GTX 1060[85][86][87][88] ... $299
12 GeForce GTX 1060[85][86][87][88] ... NaN
13 GeForce GTX 1070[89][90] ... $449 ($399)[93]
14 GeForce GTX 1070 Ti[89] ... $449
15 GeForce GTX 1080[94] ... $699 ($549)[93]
16 GeForce GTX 1080[94] ... $699 ($549)[93]
17 GeForce GTX 1080 Ti[95] ... $699
18 Nvidia TITAN X[96] ... NaN
19 Nvidia TITAN Xp[97] ... NaN
[20 rows x 25 columns],
Model ... Release price (USD)
Model ... Founders Edition
0 Nvidia TITAN V[98] ... NaN
1 Nvidia TITAN VCEO Edition[99][100] ... NaN
[2 rows x 26 columns],
Model ... Release price (USD)
Model ... Release price (USD)
0 GeForce GTX 1650[101] ... $149
1 GeForce GTX 1650[101] ... $149
2 GeForce GTX 1650 Super[103] ... $159
3 GeForce GTX 1660[106] ... $219
4 GeForce GTX 1660 Super[107] ... $229
5 GeForce GTX 1660 Ti[108] ... $279
[6 rows x 24 columns],
Model ... Release price (USD)
Model ... Founders Edition
0 GeForce RTX 2060[109] ... NaN
1 GeForce RTX 2060[109] ... NaN
2 GeForce RTX 2060 Super[111][112] ... $399
3 GeForce RTX 2070[113][114] ... $599
4 GeForce RTX 2070 Super[111][115] ... $499
5 GeForce RTX 2080[116][117] ... $799
6 GeForce RTX 2080 Super[111][118] ... $699
7 GeForce RTX 2080 Ti[119] ... $1,199
8 Nvidia TITAN RTX[120] ... $2,499
[9 rows x 28 columns],
Model ... Release price (USD)
Model ... Founders Edition
0 GeForce RTX 3060[121] ... $329
1 GeForce RTX 3060 Ti[122] ... $399
2 GeForce RTX 3070[123] ... $499
3 GeForce RTX 3080[126] ... $699
4 GeForce RTX 3090[127] ... $1499
[5 rows x 29 columns],
Model Launch ... Memory
Model Launch ... Bus type Bus width (bit)
0 GeForce2 Go 100 February 6, 2001 ... DDR 32
1 GeForce2 Go November 11, 2000 ... SDRDDR 12864
2 GeForce2 Go 200 February 6, 2001 ... DDR 64
[3 rows x 15 columns],
Model Launch ... API support
Model Launch ... Direct3D OpenGL
0 GeForce4 Go 410 February 6, 2002 ... 8.0a 1.3
1 GeForce4 Go 420 February 6, 2002 ... 8.0a 1.3
2 GeForce4 Go 440 February 6, 2002 ... 8.0a 1.3
3 GeForce4 Go 460 October 14, 2002 ... 8.0a 1.3
4 GeForce4 Go 488 NaN ... 8.0a 1.3
5 GeForce4 Go 4200 November 14, 2002 ... 8.0a 1.3
[6 rows x 17 columns],
Model Launch ... Supported API version TBP (Watts)
Model Launch ... OpenGL TBP (Watts)
Model Launch ... Drivers (Software) TBP (Watts)
0 GeForce FX Go 5100* March 2003 ... 2.1** Unknown
1 GeForce FX Go 5500* March 2003 ... 2.1** Unknown
2 GeForce FX Go 5600* March 2003 ... 2.1** Unknown
3 GeForce FX Go 5650* March 2003 ... 2.1** Unknown
4 GeForce FX Go 5700* February 1, 2005 ... 2.1** Unknown
[5 rows x 18 columns],
Model ... Memory
Model ... Bus width (bit)
0 GeForce Go 6100 + nForce Go 430 ... 64/128
1 GeForce Go 6150 + nForce Go 430 ... 64/128
2 GeForce Go 6200 ... 32
3 GeForce Go 6400 ... 64
4 GeForce Go 6600 ... 128
5 GeForce Go 6800 ... 256
6 GeForce Go 6800 Ultra ... 256
[7 rows x 16 columns],
Model ... Features
Model ... Features
0 GeForce 7000M ... NaN
1 GeForce 7150M ... NaN
2 GeForce Go 72002 ... Transparency Anti-Aliasing
3 GeForce Go 73002 ... Transparency Anti-Aliasing
4 GeForce Go 74002 ... Transparency Anti-Aliasing
5 GeForce Go 7600 ... Scalable Link Interface (SLI), Transparency An...
6 GeForce Go 7600 GT ... Scalable Link Interface (SLI), Transparency An...
7 GeForce Go 7700 ... Scalable Link Interface (SLI), Transparency An...
8 GeForce Go 7800 ... Scalable Link Interface (SLI), Transparency An...
9 GeForce Go 7800 GTX ... Scalable Link Interface (SLI), Transparency An...
10 GeForce Go 7900 GS ... Scalable Link Interface (SLI), Transparency An...
11 GeForce Go 7900 GTX ... Scalable Link Interface (SLI), Transparency An...
12 GeForce Go 7950 GTX ... Scalable Link Interface (SLI), Transparency An...
[13 rows x 18 columns],
Model ... Notes
Model ... Notes
0 GeForce 8200M G[128] ... PureVideo HD with VP3, Full H.264 / VC-1 / MPE...
1 GeForce 8400M G ... PureVideo HD with VP2, BSP Engine, and AES128 ...
2 GeForce 8400M GS ... PureVideo HD with VP2, BSP Engine, and AES128 ...
3 GeForce 8400M GT ... PureVideo HD with VP2, BSP Engine, and AES128 ...
4 GeForce 8600M GS ... PureVideo HD with VP2, BSP Engine, and AES128 ...
5 GeForce 8600M GT ... PureVideo HD with VP2, BSP Engine, and AES128 ...
6 GeForce 8700M GT ... Scalable Link Interface, PureVideo HD with VP2...
7 GeForce 8800M GTS ... Scalable Link Interface, PureVideo HD with VP2...
8 GeForce 8800M GTX ... Scalable Link Interface, PureVideo HD with VP2...
[9 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce 9100M G mGPU ... Similar to 8400M G
1 GeForce 9200M GS ... NaN
2 GeForce 9300M G ... NaN
3 GeForce 9300M GS ... NaN
4 GeForce 9400M G ... PureVideo HD with VP3. Known as the GeForce 94...
5 GeForce 9500M G ... NaN
6 GeForce 9500M GS ... Rebranded 8600M GT
7 GeForce 9600M GS ... NaN
8 GeForce 9600M GT ... NaN
9 GeForce 9650M GS ... Rebranded 8700M GT
10 GeForce 9650M GT ... NaN
11 GeForce 9700M GT ... NaN
12 GeForce 9700M GTS ... NaN
13 GeForce 9800M GS ... Down Clocked 9800M GTS Via Firmware
14 GeForce 9800M GTS ... NaN
15 GeForce 9800M GT ... Rebranded 8800M GTX
16 GeForce 9800M GTX ... NaN
17 Model ... Notes
18 Model ... Notes
[19 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce G 102M ... PureVideo HD, CUDA, Hybrid SLI, based on GeFor...
1 GeForce G 103M ... PureVideo HD, CUDA, Hybrid SLI, comparable to ...
2 GeForce G 105M ... PureVideo HD, CUDA, Hybrid SLI, comparable to ...
3 GeForce G 110M ... PureVideo HD, CUDA, Hybrid SLI
4 GeForce GT 120M ... PureVideo HD, CUDA, Hybrid SLI, Comparable to ...
5 GeForce GT 130M ... PureVideo HD, CUDA, Hybrid SLI, comparable to ...
6 GeForce GTS 150M ... PureVideo HD, CUDA, Hybrid SLI
7 GeForce GTS 160M ... PureVideo HD, CUDA, Hybrid SLI
[8 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce G210M ... Lower clocked versions of the GT218 core is al...
1 GeForce GT 220M ... rebranded 9600M GT @55 nm node shrink
2 GeForce GT 230M ... NaN
3 GeForce GT 240M ... NaN
4 GeForce GTS 250M ... NaN
5 GeForce GTS 260M ... NaN
6 GeForce GTX 260M ... NaN
7 GeForce GTX 280M ... NaN
8 GeForce GTX 285M ... Higher Clocked Version of GTX280M with new memory
[9 rows x 20 columns],
Model Launch ... Processing power (GFLOPS)2 TBP (Watts)
Model Launch ... Processing power (GFLOPS)2 TBP (Watts)
0 GeForce 305M January 10, 2010 ... 55.00 14
1 GeForce 310M January 10, 2010 ... 73.00 14
2 GeForce 315M January 5, 2011 ... 58.18 14
3 GeForce 320M April 1, 2010 ... 136.80 20
4 GeForce GT 320M January 21, 2010 ... 90.00 14
5 GeForce GT 325M January 10, 2010 ... 142.00 23
6 GeForce GT 330M January 10, 2010 ... 182.00 23
7 GeForce GT 335M January 7, 2010 ... 233.00 28?
8 GeForce GTS 350M January 7, 2010 ... 360.00 28
9 GeForce GTS 360M January 7, 2010 ... 413.00 38
[10 rows x 19 columns],
Model ... Notes
Model ... Notes
0 GeForce 410M ... Similar to Desktop GT420 OEM
1 GeForce GT 415M ... Similar to Desktop GT420 OEM
2 GeForce GT 420M ... Similar to Desktop GT430
3 GeForce GT 425M ... Similar to Desktop GT430
4 GeForce GT 435M ... Similar to Desktop GT430/440
5 GeForce GT 445M ... Similar to Desktop GTS450 OEM)
6 GeForce GTX 460M ... Similar to Desktop GTX550 Ti
7 GeForce GTX 470M ... Similar to Desktop GTX 460/560SE
8 GeForce GTX 480M ... Similar to Desktop GTX465
9 GeForce GTX 485M ... Similar to Desktop GTX560 Ti
[10 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce GT 520M ... Similar to Desktop 510/520
1 GeForce GT 520M ... Noticed in Lenovo laptops, similar to Desktop ...
2 GeForce GT 520MX ... Similar to Desktop 510 & GT520
3 GeForce GT 525M ... Similar to Desktop GT 530/430/440
4 GeForce GT 540M ... Similar to Desktop GT 530/440
5 GeForce GT 550M ... Similar to Desktop GT 530/440
6 GeForce GT 555M ... Similar to Desktop GT545
7 GeForce GTX 560M ... Similar to Desktop GTX 550Ti
8 GeForce GTX 570M[132] ... Similar to Desktop GTX 560
9 GeForce GTX 580M ... Similar to Desktop GTX 560 Ti
[10 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce 610M[133] ... OEM. Rebadged GT 520MX
1 GeForce GT 620M[134] ... OEM. Die-Shrink GF108
2 GeForce GT 625M ... OEM. Die-Shrink GF108
3 GeForce GT 630M[134][135][136] ... GF108: OEM. Rebadged GT 540MGF117: OEM Die-Shr...
4 GeForce GT 635M[134][137][138] ... GF106: OEM. Rebadged GT 555MGF116: 94% of desk...
5 GeForce GT 640M LE[134] ... GF108: 94% of desktop GT630[original research?...
6 GeForce GT 640M[134][139] ... 59% of desktop GTX650[original research?]
7 GeForce GT 645M ... 67% of desktop GTX650[original research?]
8 GeForce GT 650M[134][140][141] ... 79% of desktop GTX650[original research?]
9 GeForce GTX 660M[134][141][142][143] ... 79% of desktop GTX650[original research?]
10 GeForce GTX 670M[134] ... 73% of desktop GTX 560[original research?]
11 GeForce GTX 670MX ... 61% of desktop GTX 660[original research?]
12 GeForce GTX 675M[134] ... 75% of desktop GTX 560Ti[original research?]
13 GeForce GTX 675MX ... 61% of desktop GTX 660[original research?]
14 GeForce GTX 680M ... 78% of desktop GTX 670[original research?]
15 GeForce GTX 680MX ... 72% of desktop GTX 680[original research?]
16 Model ... Notes
17 Model ... Notes
[18 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce 710M ... OEM. About 115% of Mobile 620 & Desktop 530[or...
1 GeForce GT 720M ... OEM. About 130% of Mobile 625/630 & Desktop 62...
2 GeForce GT 720M ... Kepler, similar to 730M with half of the cores...
3 GeForce GT 730M ... Kepler, similar to Desktop GT640
4 GeForce GT 735M ... Kepler, similar to Desktop GT640
5 GeForce GT 740M ... Kepler, similar to Desktop GT640.
6 GeForce GT 740M ... about 76% of Desktop GTX650[original research?]
7 GeForce GT 745M ... about 79% of Desktop GTX650[original research?]
8 GeForce GT 750M ... about 91% of Desktop GTX650[original research?]
9 GeForce GT 755M[145] ... about 93% of Desktop GTX650[original research?]
10 GeForce GTX 760M ... about 71% of Desktop GTX 650Ti[original resear...
11 GeForce GTX 765M ... about 92% of Desktop GTX 650Ti[original resear...
12 GeForce GTX 770M ... about 83% of Desktop GTX660[original research?]
13 GeForce GTX 780M ... about 78% of Desktop GTX770[original research?]
14 Model ... Notes
15 Model ... Notes
[16 rows x 20 columns],
Model ... Notes (original research)
Model ... Notes (original research)
0 GeForce 810M ... NaN
1 GeForce 820M[146] ... 115% of 620 (Fermi)
2 GeForce 825M[148] ... 94% of 630 (Kepler)
3 GeForce 830M[149] ... 50% of 750 (Maxwell)
4 GeForce 840M[150] ... 50–80% of 745 (Maxwell)
5 GeForce 845M[151][152] ... NaN
6 GeForce 845M[151][152] ... NaN
7 GeForce GTX 850M[155] ... 80% of 750Ti
8 GeForce GTX 850M[155] ... 85% of 750Ti
9 GeForce GTX 860M[156] ... equal to 750Ti
10 GeForce GTX 860M[156] ... similar to 660 OEM.
11 GeForce GTX 870M[157] ... 105% of 660Ti
12 GeForce GTX 880M[158] ... 90% of 770
13 Model ... Notes
14 Model ... Notes
[15 rows x 20 columns],
Model ... Notes
Model ... Notes
0 GeForce 920M ... NaN
1 GeForce 930M ... NaN
2 GeForce 940M ... NaN
3 GeForce 940MX[161] ... NaN
4 GeForce GTX 950M ... Similar core config to GTX 750 Ti (GM107-400-A2)
5 GeForce GTX 950M ... Similar core config to GTX 750 Ti (GM107-400-A2)
6 GeForce GTX 960M ... Similar core config to GTX 750 Ti (GM107-400-A2)
7 GeForce GTX 965M[164] ... Similar core config to GTX 960 (GM206-300)
8 GeForce GTX 970M[166] ... Similar core config to GTX 960 OEM (GM204)
9 GeForce GTX 980M[167] ... Similar core config to GTX 970 (GM204-200) wit...
10 GeForce GTX 980[168] ... Similar to Desktop GTX 980
11 Model ... Notes
12 Model ... Notes
[13 rows x 21 columns],
Model ... SLI support
Model ... SLI support
0 GeForce MX110[169][170] ... No
1 GeForce MX130[169][171][172] ... No
2 GeForce MX150[173] ... No
3 GeForce MX150[173] ... No
4 GeForce GTX 1050 (Notebook)[176][177] ... No
5 GeForce GTX 1050 Ti (Notebook)[176] ... No
6 GeForce GTX 1060 (Notebook)[176] ... No
7 GeForce GTX 1060 Max-Q ... No
8 GeForce GTX 1060 Max-Q ... NaN
9 GeForce GTX 1070 (Notebook)[176] ... Yes
10 GeForce GTX 1070 Max-Q ... No
11 GeForce GTX 1080 (Notebook)[176] ... Yes
12 GeForce GTX 1080 Max-Q ... No
[13 rows x 26 columns],
Model Launch ... Processing power (GFLOPS) TBP (Watts)
Model Launch ... Half precision TBP (Watts)
0 GeForce GTX 1650 (Laptop) April 23, 2019 ... 6390 50
1 GeForce GTX 1650 Max-Q April 23, 2019 ... 5100 30
2 GeForce GTX 1650 Ti Max-Q April 2, 2020 ... 4915 35
3 GeForce GTX 1650 Ti April 2, 2020 ... 6083 55
4 GeForce GTX 1660 (Laptop) ? ... 8141 ?
5 GeForce GTX 1660 Ti Max-Q April 23, 2019 ... 8202 60
6 GeForce GTX 1660 Ti (Laptop)[183] April 23, 2019 ... 9769 80
[7 rows x 26 columns],
Model ... TDP (Watts)
Model ... TDP (Watts)
0 GeForce MX230[186] ... 10
1 GeForce MX250[187][188] ... 25
2 GeForce MX250[187][188] ... 10
3 GeForce MX330[189] ... 25
4 GeForce MX350[190] ... 20
5 GeForce MX450[191][192] ... 25
6 GeForce MX450[191][192] ... 25
7 GeForce RTX 2060[193] ... 80
8 GeForce RTX 2060 Max-Q[194] ... 65
9 GeForce RTX 2070[195] ... 115
10 GeForce RTX 2070 Max-Q[196] ... 80
11 GeForce RTX 2070 Super[197] ... 115
12 GeForce RTX 2070 Super Max-Q[198] ... 80
13 GeForce RTX 2080[199] ... 150
14 GeForce RTX 2080 Max-Q[200] ... 80
15 GeForce RTX 2080 Super[201] ... 150
16 GeForce RTX 2080 Super Max-Q[202] ... 80
[17 rows x 25 columns],
Model ... TDP (Watts)
Model ... TDP (Watts)
0 GeForce RTX 3050[203] ... NaN
1 GeForce RTX 3060[204] ... NaN
2 GeForce RTX 3060 Max Q[205] ... NaN
3 GeForce RTX 3070[206][207] ... 115.0
4 GeForce RTX 3070 Max Q[208] ... NaN
5 GeForce RTX 3080[209][210] ... 115.0
[6 rows x 25 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro NV10GL ... NaN NaN
1 Quadro DDR NV10GL ... NaN NaN
2 Quadro2 MXR NV11GL ... NaN NaN
3 Quadro2 EX NV11GL ... NaN NaN
4 Quadro2 PRO NV15GL ... NaN NaN
5 Quadro DCC NV20GL ... NaN NaN
6 Quadro4 380XGL NV18GL ... NaN NaN
7 Quadro4 500XGL NV17GL ... NaN NaN
8 Quadro4 550XGL NV17GL ... NaN NaN
9 Quadro4 580XGL NV18GL ... NaN NaN
10 Quadro4 700XGL NV25 ... NaN NaN
11 Quadro4 750XGL NV25 ... NaN Stereo display
12 Quadro4 900XGL NV25 ... NaN Stereo display
13 Quadro4 980XGL NV28GL ... NaN Stereo display
14 Model Code name ... TBP (Watts) Features
15 Model Code name ... TBP (Watts) Features
[16 rows x 17 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro FX 500 NV34GL ... NaN Stereo display
1 Quadro FX 600 NV34GL ... NaN Stereo display
2 Quadro FX 700 NV35GL ... NaN Stereo display
3 Quadro FX 1000 NV30GL ... NaN Stereo display
4 Quadro FX 1100 NV36GL ... NaN Stereo display
5 Quadro FX 2000 NV30GL ... NaN Stereo display
6 Quadro FX 3000 NV35GL ... NaN Stereo display
7 Quadro FX 3000G NV35GL ... NaN Stereo display, Genlock
8 Quadro FX 4000 NV40GL ... 142.0 Stereo display
9 Quadro FX 4000 SDI NV40GL ... NaN Stereo display, Genlock
[10 rows x 17 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts)
Model Code name Fab (nm) ... Direct3D OpenGL TBP (Watts)
0 Quadro FX 330 NV37GL 150 ... 9.0 2.0 21
1 Quadro FX 1300 NV38 130 ... 9.0 2.1 55
[2 rows x 16 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro FX 540 NV43GL ... 35 NaN
1 Quadro FX 1400 NV41 ... 75 Stereo display, SLI
2 Quadro FX 3400 NV45GL/ NV40 ... 101 Stereo display, SLI
3 Quadro FX 3450 NV41 ... 83 Stereo display, SLI
4 Quadro FX 4400 NV45GL A3/ NV40 ... 110 Stereo display, SLI
[5 rows x 17 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro FX 350 G72 ... 21 NaN
1 Quadro FX 550 NV43 ... 30 NaN
2 Quadro FX 1500 G71 ... 65 NaN
3 Quadro FX 3500 G71 ... 80 Stereo display, SLI
4 Quadro FX 4500 G70 ... 109 Stereo display, SLI
5 Quadro FX 4500 SDI G70 ... 116 Stereo display, Genlock
6 Quadro FX 4500X2 G71 ... 145 Stereo display, SLI, Genlock
7 Quadro FX 5500 G71 ... 96 Stereo display, SLI, Genlock
8 Quadro FX 5500 SDI G71 ... 104 Stereo display, SLI, Genlock
[9 rows x 17 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro FX 560 G73GL ... 30 NaN
1 Quadro FX 46002 G80 ... 134 Stereo display, SLI, Genlock
2 Quadro FX 4600 SDI2 G80 ... 154 Stereo display, SLI, Genlock
3 Quadro FX 56002 G80 ... 171 Stereo display, SLI, Genlock
[4 rows x 20 columns],
Model ... Notes
Model ... Notes
0 Quadro FX 370 ... NaN
1 Quadro FX 370 LP ... DMS-59 for two Single Link DVI
2 Quadro FX 470 ... based on GeForce 9400 mGPU
3 Quadro FX 570 ... NaN
4 Quadro FX 1700 ... NaN
5 Quadro FX 3700 ... Stereo display, SLI
6 Quadro FX 4700X2 ... NaN
7 Quadro VX 200 ... 2× Dual-link DVI, S-Video, optimised for Autod...
[8 rows x 20 columns],
Model ... Notes
Model ... Notes
0 Quadro FX 380 ... Two Dual Link DVI, no DisplayPort
1 Quadro FX 380 LP ... DisplayPort, Dual Link DVI
2 Quadro FX 580 ... Dual DisplayPort, Dual Link DVI
3 Quadro FX 1800 ... Stereo DP Dual Link DVI, Dual DisplayPort, SLI
4 Quadro FX 3800 ... Stereo DP Dual Link DVI, Dual DisplayPort, SLI
5 Quadro FX 4800 ... Stereo DP Dual Link DVI, Dual DisplayPort, SLI
6 Quadro FX 5800 ... Stereo DP Dual Link DVI, Dual DisplayPort, SLI
7 Quadro CX[214] ... Display Port and dual-link DVI Output, optimis...
[8 rows x 20 columns],
Model ... Notes
Model ... Notes
0 Quadro 400 ... DisplayPort, Dual Link DVI
1 Quadro 600 ... DisplayPort, Dual Link DVI
2 Quadro 2000 ... Stereo DP Dual Link DVI, Dual DisplayPort
3 Quadro 4000 ... NaN
4 Quadro 5000 ... NaN
5 Quadro 6000 ... NaN
6 Quadro 7000 ... NaN
7 Quadro Plex 7000[215] ... NaN
[8 rows x 20 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts) Notes
Model Code name Fab (nm) ... CUDA TBP (Watts) Notes
0 Quadro 410 GK107 28 ... 3.0 38 NaN
1 Quadro K600 GK107 28 ... 3.0 41 6.3" Card
2 Quadro K2000 GK107 28 ... 3.0 51 7.97" Card
3 Quadro K2000D GK107 28 ... 3.0 51 7.97" Card
4 Quadro K4000 GK106 28 ... 3.0 80 9.5" Card
5 Quadro K5000 GK104 28 ... 3.0 122 10.5" Card
6 Quadro K6000 GK110 28 ... 3.5 225 10.5" Card
[7 rows x 21 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts) Notes
Model Code name Fab (nm) ... CUDA TBP (Watts) Notes
0 Quadro K420 GK107 28 ... 3.0 41 NaN
1 Quadro K620 GM107-850 28 ... 5.0 45 6.3" Card
2 Quadro K1200 GM107-860 28 ... 5.0 45 7.97" Card
3 Quadro K2200 GM107-875-A2[218] 28 ... 5.0 68 7.97" Card
4 Quadro K4200 GK104 28 ... 3.0 105 9.5" Card
5 Quadro K5200 GK110B 28 ... 3.5 150 10.5" Card
[6 rows x 21 columns],
Model ... Notes
Model ... Notes
0 Quadro M2000 ... Four DisplayPort 1.2a
1 Quadro M4000 ... Four DisplayPort 1.2a
2 Quadro M5000 ... Four DisplayPort 1.2a, One DVI-I
3 Quadro M6000 ... Four DisplayPort 1.2a, One DVI-I
[4 rows x 23 columns],
Model ... Notes
Model ... Notes
0 Quadro P400 ... Three Mini-DisplayPort 1.4
1 Quadro P600 ... Four Mini-DisplayPort 1.4
2 Quadro P620 ... Four Mini-DisplayPort 1.4
3 Quadro P1000 ... Four Mini-DisplayPort 1.4
4 Quadro P2000 ... Four DisplayPort 1.4
5 Quadro P2200 ... Four DisplayPort 1.4
6 Quadro P4000 ... Four DisplayPort 1.4
7 Quadro P5000 ... Four DisplayPort 1.4, One DVI-D
8 Quadro P6000 ... Four DisplayPort 1.4, One DVI-D
9 Quadro GP100 ... NVLINK support
[10 rows x 23 columns],
Model Code name[178] ... TBP (Watts) Notes
Model Code name[178] ... TBP (Watts) Notes
0 Quadro GV100[226] GV100-875-A1 ... 250 4x DisplayPort, NVLINK support
[1 rows x 23 columns],
Model Code name[178] ... TBP (Watts) Notes
Model Code name[178] ... TBP (Watts) Notes
0 Quadro RTX 4000 TU104-850-A1 ... 160 3x DisplayPort 1x USB Type-C
1 Quadro RTX 5000 TU104-875-A1 ... 200 3x DisplayPort 1x USB Type-C
2 Quadro RTX 6000 TU102-875-A1 ... 250 3x DisplayPort 1x USB Type-C
3 Quadro RTX 8000 TU102-875-A1 ... 250 3x DisplayPort 1x USB Type-C
[4 rows x 23 columns],
Model ... Notes
Model ... Notes
0 NVS 50 ... DVI-I, S-Video
1 NVS 100 ... 2× DVI-I, VGA, S-Video
2 NVS 200 ... LFH-60
3 NVS 210S ... DVI + VGA
4 NVS 280 ... DMS-59
5 NVS 285 ... DMS-59
6 NVS 290 ... DMS-59
7 NVS 295 ... 2× DisplayPort or 2× DVI-D
8 NVS 300 ... DMS-59
9 NVS 310 ... 2× DisplayPort
10 NVS 315 ... DMS-59 Idle Power Consumption 7W
11 NVS 400 ... DMS-59
12 NVS 420 ... through VHDCI to (4× DisplayPort or 4× DVI-D)
13 NVS 440 ... 2× DMS-59[227]
14 NVS 450 ... 4× DisplayPort
15 NVS 510 ... 4× miniDisplayPort
16 NVS 810 ... 8× miniDisplayPort
[17 rows x 19 columns],
Model ... Notes
Model ... Notes
0 Quadro2 Go[228] ... First mobile Quadro based on the GeForce2 Go, ...
1 Quadro4 500 Go GL[231][232] ... Based on the GeForce4 Go, dynamic core clock 6...
2 Quadro4 700 Go GL[231][233][234] ... Based on GeForce4 Go 4200, uses DDR according ...
3 Quadro FX Go 700[235] ... Based on the Geforce FX Go (FX 56XX)
4 Quadro FX Go 1000[236] ... Seems to be an overclocked Quadro FX 700 Go.
5 Quadro FX Go 1400[236] ... Last chip designated as a Quadro FX Go, uses P...
[6 rows x 17 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts)
Model Code name Fab (nm) ... Direct3D OpenGL TBP (Watts)
0 Quadro FX 350M G72GLM 90 ... 9.0c 2.1 15
1 Quadro FX 1500M G71GLM 90 ... 9.0c 2.1 15
2 Quadro FX 2500M G71GLM 90 ... 9.0c 2.1 15
3 Quadro FX 3500M G71GLM 90 ... 9.0c 2.1 15
[4 rows x 16 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro FX 360M G86M ... 17 Based on the GeForce 8400M GS
1 Quadro FX 560M G73GLM ... 35? 7600GS based?
2 Quadro FX 1600M G84M ... 50? NaN
3 Quadro FX 3600M G92M ... 70 Based on the GeForce 8800M GTX
[4 rows x 19 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts)
Model Code name Fab (nm) ... Direct3D OpenGL TBP (Watts)
0 Quadro FX 370M G98M 65 ... 10.0 3.3 20
1 Quadro FX 570M G84M 80 ... 10.0 3.3 45
2 Quadro FX 770M G96M 65 ... 10.0 3.3 35
3 Quadro FX 1700M G96M 65 ... 10.0 3.3 50
4 Quadro FX 2700M G94M 65 ... 10.0 3.3 65
5 Quadro FX 3700M G92M 65 ... 10.0 3.3 75
[6 rows x 18 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts)
Model Code name Fab (nm) ... Direct3D OpenGL TBP (Watts)
0 Quadro FX 380M GT218M 40 ... 10.1 3.3 25
1 Quadro FX 880M GT216M 40 ... 10.1 3.3 35
2 Quadro FX 1800M GT215M 40 ... 10.1 3.3 45
3 Quadro FX 2800M G92M 55 ... 10.0 3.3 75
4 Quadro FX 3800M G92M 55 ... 10.0 3.3 100
[5 rows x 18 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts) Notes
Model Code name Fab (nm) ... OpenGL TBP (Watts) Notes
0 Quadro 500M GF108 40 ... 4.5 35 NaN
1 Quadro 1000M GF108 40 ... 4.5 45 Dell Precision M4600
2 Quadro 2000M GF106 40 ... 4.5 55 Dell Precision M4600
3 Quadro 3000M GF104 40 ... 4.5 75 Dell Precision M6600
4 Quadro 4000M GF104 40 ... 4.5 100 Dell Precision M6600
5 Quadro 5000M GF100 40 ... 4.5 100 Dell Precision M6500
6 Quadro 5010M GF110 40 ... 4.5 100 Dell Precision M6600
[7 rows x 19 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro K500M GK107 ... 35 NaN
1 Quadro K1000M GK107 ... 45 Dell Precision M4700
2 Quadro K2000M GK107 ... 55 Dell Precision M4700
3 Quadro K3000M GK104 ... 75 Dell Precision M6700
4 Quadro K4000M GK104 ... 100 Dell Precision M6700
5 Quadro K5000M GK104 ... 100 Dell Precision M6700
[6 rows x 20 columns],
Model Code name ... TBP (Watts) Notes
Model Code name ... TBP (Watts) Notes
0 Quadro K510M GK208 ... 30 NaN
1 Quadro K610M GK208 ... 30 NaN
2 Quadro K1100M GK107 ... 45 Dell Precision M3800 and M4800
3 Quadro K2100M GK106 ... 55 Dell Precision M4800
4 Quadro K3100M GK104 ... 75 Dell Precision M6800
5 Quadro K4100M GK104 ... 100 Dell Precision M6800
6 Quadro K5100M GK104 ... 100 Dell Precision M6800
[7 rows x 20 columns],
Model Code name Fab (nm) ... Supported API version Nvidia Optimustechnology TBP (Watts)
Model Code name Fab (nm) ... OpenGL Nvidia Optimustechnology TBP (Watts)
0 Quadro K2200M GM107 28 ... 4.5 Yes 65
[1 rows x 20 columns],
Model Code name Fab (nm) ... Supported API version Nvidia Optimustechnology TBP (Watts)
Model Code name Fab (nm) ... Vulkan Nvidia Optimustechnology TBP (Watts)
0 Quadro M500M GM108 28 ... 1.0 Yes 25
1 Quadro M600M GM107 28 ... 1.0 Yes 30
2 Quadro M1000M GM107 28 ... 1.0 Yes 40
3 Quadro M2000M GM107 28 ... 1.0 Yes 55
4 Quadro M3000M GM204 28 ... 1.0 Yes 75
5 Quadro M4000M GM204 28 ... 1.0 Yes 100
6 Quadro M5000M GM204 28 ... 1.0 Yes 100
[7 rows x 21 columns],
Model Code name Fab (nm) ... Supported API version Nvidia Optimustechnology TBP (Watts)
Model Code name Fab (nm) ... OpenGL Nvidia Optimustechnology TBP (Watts)
0 Quadro M520 GM108 28 ... 4.5 Yes 25
1 Quadro M620 GM107 28 ... 4.5 Yes 30
2 Quadro M1200 GM107 28 ... 4.5 Yes 45
3 Quadro M2200 GM206 28 ... 4.5 Yes 55
[4 rows x 19 columns],
Model Code name Fab (nm) ... Supported API version Nvidia Optimustechnology TBP (Watts)
Model Code name Fab (nm) ... OpenGL Nvidia Optimustechnology TBP (Watts)
0 Quadro M5500 GM204 28 ... 4.5 Yes 150
[1 rows x 19 columns],
Model Code name ... Nvidia Optimustechnology TBP (Watts)
Model Code name ... Nvidia Optimustechnology TBP (Watts)
0 Quadro P500 GP108 ... Yes 18
1 Quadro P600 GP107 ... Yes 25
2 Quadro P1000 GP107(N18P-Q1-A1) ... Yes 40
3 Quadro P2000 GP107(N18P-Q3-A1) ... Yes 50
4 Quadro P3000 GP106 ... Yes 75
5 Quadro P4000 GP104 ... Yes 100
6 Quadro P4000 Max-Q GP104 ... Yes 80
7 Quadro P5000 GP104 ... Yes 100
[8 rows x 19 columns],
Model Code name Fab (nm) ... Supported API version Nvidia Optimustechnology TBP (Watts)
Model Code name Fab (nm) ... OpenGL Nvidia Optimustechnology TBP (Watts)
0 Quadro P3200 GP104 16 ... 4.5 Yes 78
1 Quadro P4200 GP104 16 ... 4.5 Yes 115
2 Quadro P5200 GP104 16 ... 4.5 Yes 150
[3 rows x 19 columns],
Model Code name[178] ... TBP (Watts) Notes
Model Code name[178] ... TBP (Watts) Notes
0 Quadro T1000[237] TU117(N19P-Q1-A1) ... 40-50 NaN
1 Quadro T2000[237] TU117(N19P-Q3-A1) ... 60 NaN
2 Quadro T2000 Max-Q TU117 ... 40 NaN
3 Quadro RTX 3000[237] TU106(N19E-Q1-KA-K1) ... 60-80 NaN
4 Quadro RTX 3000 Max-Q[238] TU106 ... 60 NaN
5 Quadro RTX 4000[237] TU104(N19E-Q3-A1) ... 110 NaN
6 Quadro RTX 4000 Max-Q[239] TU104 ... 80 NaN
7 Quadro RTX 5000[237] TU104(N19E-Q5-A1) ... 110 NaN
8 Quadro RTX 5000 Max-Q[240] TU104 ... 80 NaN
[9 rows x 25 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts) Notes
Model Code name Fab (nm) ... OpenGL TBP (Watts) Notes
0 Quadro NVS 110M G72M 90 ... 2.1 10 NaN
1 Quadro NVS 120M G72GLM 90 ... 2.1 10 NaN
2 Quadro NVS 130M G86M 80 ... 3.3 10 NaN
3 Quadro NVS 135M G86M 80 ... 3.3 10 NaN
4 Quadro NVS 140M G86M 80 ... 3.3 10 NaN
5 Quadro NVS 150M G98M 65 ... 3.3 10 NaN
6 Quadro NVS 160M G98M 65 ... 3.3 12 NaN
7 Quadro NVS 300M G72GLM 90 ... 2.1 16 NaN
8 Quadro NVS 320M G84M 65 ... 3.3 20 NaN
9 Quadro NVS 510M G72GLM 90 ... 2.1 45? based on Go 7900 GTX
[10 rows x 19 columns],
Model Code name Fab (nm) ... Supported API version TBP (Watts) Notes
Model Code name Fab (nm) ... OpenGL TBP (Watts) Notes
0 NVS 2100M GT218M 40 ... 3.3 14.0 NaN
1 NVS 3100M GT218M 40 ... 3.3 14.0 based on G210M/310M
2 NVS 4200M GF119 40 ... 4.5 NaN based on GT 520M
3 NVS 5100M GT216M 40 ... 3.3 35.0 NaN
4 NVS 5200M GF108 40/28 ... 4.5 35.0 NaN
5 NVS 5400M GF108 40/28 ... 4.5 35.0 NaN
[6 rows x 19 columns],
Model Archi-tecture Chips ... Memory TDP (Watts)
Model Archi-tecture Chips ... Bus type Size (GB) TDP (Watts)
0 GRID K1[241] Kepler 4× GK107 ... DDR3 4× 4GB 130
1 GRID K2[242] Kepler 2× GK104-895 ... GDDR5 2× 4GB 225
2 GRID K340 Kepler 4× GK107 ... GDDR5 4× 1GB 225
3 GRID K520 Kepler 2× GK104 ... GDDR5 2× 4GB 225
[4 rows x 8 columns],
Model ... Notes, form_factor
Model ... Notes, form_factor
Units ... Unnamed: 18_level_2
0 C870 GPU Computing Module[h] ... Internal PCIe GPU (full-height, dual-slot)
1 D870 Deskside Computer[h] ... Deskside or 3U rack-mount external GPUs
2 S870 GPU Computing Server[h] ... 1U rack-mount external GPUs, connect via 2× PC...
3 C1060 GPU Computing Module[i] ... Internal PCIe GPU (full-height, dual-slot)
4 S1070 GPU Computing Server "400 configuration"[i] ... 1U rack-mount external GPUs, connect via 2× PC...
5 S1070 GPU Computing Server "500 configuration"[i] ... 1U rack-mount external GPUs, connect via 2× PC...
6 S1075 GPU Computing Server[i][246] ... 1U rack-mount external GPUs, connect via 1× PC...
7 Quadro Plex 2200 D2 Visual Computing System[j] ... Deskside or 3U rack-mount external GPUs with 4...
8 Quadro Plex 2200 S4 Visual Computing System[j] ... 1U rack-mount external GPUs, connect via 2× PC...
9 C2050 GPU Computing Module[247] ... Internal PCIe GPU (full-height, dual-slot)
10 M2050 GPU Computing Module[248] ... Internal PCIe GPU (full-height, dual-slot)
11 C2070 GPU Computing Module[247] ... Internal PCIe GPU (full-height, dual-slot)
12 C2075 GPU Computing Module[249] ... Internal PCIe GPU (full-height, dual-slot)
13 M2070/M2070Q GPU Computing Module[250] ... Internal PCIe GPU (full-height, dual-slot)
14 M2090 GPU Computing Module[251] ... Internal PCIe GPU (full-height, dual-slot)
15 S2050 GPU Computing Server ... 1U rack-mount external GPUs, connect via 2× PC...
16 S2070 GPU Computing Server ... 1U rack-mount external GPUs, connect via 2× PC...
17 K10 GPU accelerator[252] ... Internal PCIe GPU (full-height, dual-slot)
18 K20 GPU accelerator[253][254] ... Internal PCIe GPU (full-height, dual-slot)
19 K20X GPU accelerator[255] ... Internal PCIe GPU (full-height, dual-slot)
20 K40 GPU accelerator[256] ... Internal PCIe GPU (full-height, dual-slot)
21 K80 GPU accelerator[257] ... Internal PCIe GPU (full-height, dual-slot)
22 M4 GPU accelerator[258][259] ... Internal PCIe GPU (half-height, single-slot)
23 M6 GPU accelerator[260] ... Internal MXM GPU
24 M10 GPU accelerator[261] ... Internal PCIe GPU (full-height, dual-slot)
25 M40 GPU accelerator[259][262] ... Internal PCIe GPU (full-height, dual-slot)
26 M60 GPU accelerator[263] ... Internal PCIe GPU (full-height, dual-slot)
27 P4 GPU accelerator[264] ... PCIe card
28 P6 GPU accelerator[265][266] ... MXM card
29 P40 GPU accelerator[264] ... PCIe card
30 P100 GPU accelerator (mezzanine)[267][268] ... NVLink card
31 P100 GPU accelerator (16 GB card)[269] ... PCIe card
32 P100 GPU accelerator (12 GB card)[269] ... PCIe card
33 V100 GPU accelerator (mezzanine)[270][271][272] ... NVlink card
34 V100 GPU accelerator (PCIe card)[270][271][272] ... PCIe card
35 T4 GPU accelerator (PCIe card)[273][274] ... PCIe card
36 A100 GPU accelerator (PCIe card)[275][276] ... PCIe card
37 Model ... Notes, form factor
38 Model ... Notes, form factor
[39 rows x 19 columns],
Model ... API support
Model ... Other
0 XGPU (Xbox)[277][278] ... NaN
1 RSX (PS3)[279][280][281] ... NaN
2 NX-SoC (Nintendo Switch)[282][283] ... Vulkan 1.0[286]NVN[282]
[3 rows x 20 columns],
.mw-parser-output .navbar{display:inline;font-size:88%;font-weight:normal}.mw-parser-output .navbar-collapse{float:left;text-align:left}.mw-parser-output .navbar-boxtext{word-spacing:0}.mw-parser-output .navbar ul{display:inline-block;white-space:nowrap;line-height:inherit}.mw-parser-output .navbar-brackets::before{margin-right:-0.125em;content:"[ "}.mw-parser-output .navbar-brackets::after{margin-left:-0.125em;content:" ]"}.mw-parser-output .navbar li{word-spacing:-0.125em}.mw-parser-output .navbar-mini abbr{font-variant:small-caps;border-bottom:none;text-decoration:none;cursor:inherit}.mw-parser-output .navbar-ct-full{font-size:114%;margin:0 7em}.mw-parser-output .navbar-ct-mini{font-size:114%;margin:0 4em}.mw-parser-output .infobox .navbar{font-size:100%}.mw-parser-output .navbox .navbar{display:block;font-size:100%}.mw-parser-output .navbox-title .navbar{float:left;text-align:left;margin-right:0.5em}vteNvidia .mw-parser-output .navbar{display:inline;font-size:88%;font-weight:normal}.mw-parser-output .navbar-collapse{float:left;text-align:left}.mw-parser-output .navbar-boxtext{word-spacing:0}.mw-parser-output .navbar ul{display:inline-block;white-space:nowrap;line-height:inherit}.mw-parser-output .navbar-brackets::before{margin-right:-0.125em;content:"[ "}.mw-parser-output .navbar-brackets::after{margin-left:-0.125em;content:" ]"}.mw-parser-output .navbar li{word-spacing:-0.125em}.mw-parser-output .navbar-mini abbr{font-variant:small-caps;border-bottom:none;text-decoration:none;cursor:inherit}.mw-parser-output .navbar-ct-full{font-size:114%;margin:0 7em}.mw-parser-output .navbar-ct-mini{font-size:114%;margin:0 4em}.mw-parser-output .infobox .navbar{font-size:100%}.mw-parser-output .navbox .navbar{display:block;font-size:100%}.mw-parser-output .navbox-title .navbar{float:left;text-align:left;margin-right:0.5em}vteNvidia.1
0 GeForce (List of GPUs)Fixed pixel pipelinePre-... GeForce (List of GPUs)Fixed pixel pipelinePre-...
1 GeForce (List of GPUs) GeForce (List of GPUs)
2 Fixed pixel pipelinePre-GeForce NV1 NV2 RIVA ... Fixed pixel pipelinePre-GeForce NV1 NV2 RIVA ...
3 Fixed pixel pipeline Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2 G...
4 Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2
5 Vertex and pixel shaders GeForce 3 4 Ti FX 6 7
6 Unified shaders GeForce 8 9 100 200 300 400 500
7 Unified shaders & NUMA GeForce 600 700 800M 900 10 16
8 Ray tracing GeForce 20 30
9 Software and technologiesMultimedia accelerati... Software and technologiesMultimedia accelerati...
10 Software and technologies Software and technologies
11 Multimedia acceleration NVENC (video encoding)... Multimedia acceleration NVENC (video encoding)...
12 Multimedia acceleration NVENC (video encoding) NVDEC (video decoding) ...
13 Software Cg (shading language) CUDA Gelato (offline ren...
14 Technologies Nvidia 3D Vision (stereo 3D) Nvidia G-Sync (va...
15 GPU microarchitectures Tesla Fermi Kepler Maxwell Pascal Volta Turing...
16 Other productsGraphics processing Nvidia Quadr... Other productsGraphics processing Nvidia Quadr...
17 Other products Other products
18 Graphics processing Nvidia Quadro Quadro Plex ... Graphics processing Nvidia Quadro Quadro Plex ...
19 Graphics processing Nvidia Quadro Quadro Plex
20 GPGPU Nvidia Tesla DGX
21 Console components NV2A (Xbox) RSX 'Reality Synthesizer' (PlaySta...
22 Nvidia Shield Shield Portable Shield Tablet Shield Android T...
23 SoCs and embedded GoForce Drive Jetson Tegra
24 CPUs Project Denver
25 Computer chipsets nForce
26 CompanyKey people Jen-Hsun Huang Chris Malacho... CompanyKey people Jen-Hsun Huang Chris Malacho...
27 Company Company
28 Key people Jen-Hsun Huang Chris Malachowsky Cu... Key people Jen-Hsun Huang Chris Malachowsky Cu...
29 Key people Jen-Hsun Huang Chris Malachowsky Curtis Priem ...
30 Acquisitions 3dfx Interactive Ageia ULi Arm Holdings Icera ... ,
GeForce (List of GPUs) GeForce (List of GPUs).1
0 Fixed pixel pipelinePre-GeForce NV1 NV2 RIVA ... Fixed pixel pipelinePre-GeForce NV1 NV2 RIVA ...
1 Fixed pixel pipeline Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2 G...
2 Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2
3 Vertex and pixel shaders GeForce 3 4 Ti FX 6 7
4 Unified shaders GeForce 8 9 100 200 300 400 500
5 Unified shaders & NUMA GeForce 600 700 800M 900 10 16
6 Ray tracing GeForce 20 30,
0 1
0 Fixed pixel pipeline Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2 G...
1 Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2
2 Vertex and pixel shaders GeForce 3 4 Ti FX 6 7
3 Unified shaders GeForce 8 9 100 200 300 400 500
4 Unified shaders & NUMA GeForce 600 700 800M 900 10 16
5 Ray tracing GeForce 20 30,
0 1
0 Pre-GeForce NV1 NV2 RIVA 128 RIVA TNT TNT2,
Software and technologies Software and technologies.1
0 Multimedia acceleration NVENC (video encoding)... Multimedia acceleration NVENC (video encoding)...
1 Multimedia acceleration NVENC (video encoding) NVDEC (video decoding) ...
2 Software Cg (shading language) CUDA Gelato (offline ren...
3 Technologies Nvidia 3D Vision (stereo 3D) Nvidia G-Sync (va...
4 GPU microarchitectures Tesla Fermi Kepler Maxwell Pascal Volta Turing...,
0 1
0 Multimedia acceleration NVENC (video encoding) NVDEC (video decoding) ...
1 Software Cg (shading language) CUDA Gelato (offline ren...
2 Technologies Nvidia 3D Vision (stereo 3D) Nvidia G-Sync (va...
3 GPU microarchitectures Tesla Fermi Kepler Maxwell Pascal Volta Turing...,
Other products Other products.1
0 Graphics processing Nvidia Quadro Quadro Plex ... Graphics processing Nvidia Quadro Quadro Plex ...
1 Graphics processing Nvidia Quadro Quadro Plex
2 GPGPU Nvidia Tesla DGX
3 Console components NV2A (Xbox) RSX 'Reality Synthesizer' (PlaySta...
4 Nvidia Shield Shield Portable Shield Tablet Shield Android T...
5 SoCs and embedded GoForce Drive Jetson Tegra
6 CPUs Project Denver
7 Computer chipsets nForce,
0 1
0 Graphics processing Nvidia Quadro Quadro Plex
1 GPGPU Nvidia Tesla DGX
2 Console components NV2A (Xbox) RSX 'Reality Synthesizer' (PlaySta...
3 Nvidia Shield Shield Portable Shield Tablet Shield Android T...
4 SoCs and embedded GoForce Drive Jetson Tegra
5 CPUs Project Denver
6 Computer chipsets nForce,
Company Company.1
0 Key people Jen-Hsun Huang Chris Malachowsky Cu... Key people Jen-Hsun Huang Chris Malachowsky Cu...
1 Key people Jen-Hsun Huang Chris Malachowsky Curtis Priem ...
2 Acquisitions 3dfx Interactive Ageia ULi Arm Holdings Icera ...,
0 1
0 Key people Jen-Hsun Huang Chris Malachowsky Curtis Priem ...
1 Acquisitions 3dfx Interactive Ageia ULi Arm Holdings Icera ...,
vteGraphics processing unit vteGraphics processing unit.1
0 GPU Desktop Intel Xe GT Nvidia GeForce Quadro Tesl...
1 Desktop Intel Xe GT Nvidia GeForce Quadro Tesla Tegra ...
2 Mobile Adreno Apple Mali PowerVR VideoCore Vivante Im...
3 Architecture Compute kernel Fabrication CMOS FinFET MOSFET ...
4 Components Blitter Geometry processor Input–output memory...
5 Memory DMA Framebuffer SGRAM GDDR GDDR3 GDDR4 GDDR5 G...
6 Form factor IP core Discrete graphics Clustering Switching...
7 Performance Clock rate Display resolution Fillrate Pixel/s...
8 Misc 2D Scrolling Sprite Tile 3D GI Texture ASIC GP...,
0 1
0 Desktop Intel Xe GT Nvidia GeForce Quadro Tesla Tegra ...
1 Mobile Adreno Apple Mali PowerVR VideoCore Vivante Im...]What happened there?
We only passed an argument, the url from which to extract a table. With that information alone pandas.read_html will return you a list of all the tables it was able to extract from the page.
pandas.read_html() returns a list of dataframes, one for each table it finds in the page.
It seems to have a lot of tables in this wikipedia page. Let’s count how many.
print(len(tables))97That’s far too many tables for what we need. In this case we only want to extract the table which contains information about the NVIDIA RTX 30 series graphics cards.
Upon inspection the HTML of the table shows that there are no unique id or css classes that we can make use of. However that shouldn’t stop us.
if we pass in the parameter match we should be able to pass in a string with a distinct string in the table so it can find the table that I want.
I will pick a string that shows the CUDA core config for the RTX3090. This should allow me to only get the table I want.
pd = pandas.read_html("https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units", match="10496:328:112:328:82")
print(len(pd))
1display(pd[0])| Model | Launch | Code name | Process | Transistors (billion) | Die size (mm2) | Core config[a] | Bus interface | L2 Cache(MB) | Clock speeds | Memory | Fillrate | Processing power (TFLOPS) | Ray-tracing Performance | TDP (Watts) | NVLink support | Release price (USD) | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Model | Launch | Code name | Process | Transistors (billion) | Die size (mm2) | Core config[a] | Bus interface | L2 Cache(MB) | Base core clock (MHz) | Boost core clock (MHz) | Memory (MT/s) | Size (GiB) | Bandwidth (GB/s) | Bus type | Bus width (bit) | Pixel (GP/s) | Texture (GT/s) | Single precision | Double precision | Half precision | Tensor compute (FP16) (2:1 sparse) | Rays/s (Billions) | RTX OPS/s (Trillions) | Ray Perf TFLOPS | TDP (Watts) | NVLink support | MSRP | Founders Edition | |
| 0 | GeForce RTX 3060[121] | February 25, 2021 | GA106-300-A1 | Samsung8N(10LP++++) | 13.25 | 300.0 | 3584:112:64:112:28(28) (?) | PCIe 4.0 x16 | 3 | 1320 | 1777 | 15000 | 12 | 360.0 | GDDR6 | 192 | NaN | NaN | 9.46 12.74 | 0.148 0.199 | 9.46 12.74 | NaN | NaN | NaN | 25 | 170 | No | NaN | $329 |
| 1 | GeForce RTX 3060 Ti[122] | December 2, 2020 | GA104-200-A1 | Samsung8N(10LP++++) | 17.40 | 392.5 | 4864:152:80:152:38(38) (6) | PCIe 4.0 x16 | 4 | 1410 | 1665 | 14000 | 8 | 448.0 | GDDR6 | 256 | NaN | NaN | 13.7 16.20 | 0.214 0.253 | 13.7 16.20 | ? 129.6 | NaN | NaN | NaN | 200 | No | NaN | $399 |
| 2 | GeForce RTX 3070[123] | October 29, 2020[124] | GA104-300-A1 | Samsung8N(10LP++++) | 17.40 | 392.5 | 5888:184:96:184:46(46) (6) | PCIe 4.0 x16 | 4 | 1500 | 1725 | 14000 | 8 | 448.0 | GDDR6 | 256 | 96.0 110.72 | 276.0 318.32 | 17.66 20.37 | 0.276 0.318 | 17.66 20.37 | 141.31 162.98 | NaN | NaN | 40[125] | 220 | No | NaN | $499 |
| 3 | GeForce RTX 3080[126] | September 17, 2020 | GA102-200-KD-A1 | Samsung8N(10LP++++) | 28.30 | 628.4 | 8704:272:96:272:68(68) (7) | PCIe 4.0 x16 | 5 | 1440 | 1710 | 19000 | 10 | 760.0 | GDDR6X | 320 | 126.72 150.48 | 391.68 465.12 | 25.06 29.76 | 0.392 0.465 | 25.06 29.76 | 200.54 238.14 | NaN | NaN | 58[125] | 320 | No | NaN | $699 |
| 4 | GeForce RTX 3090[127] | September 24, 2020 | GA102-300-A1 | Samsung8N(10LP++++) | 28.30 | 628.4 | 10496:328:112:328:82(82) (7) | PCIe 4.0 x16 | 6 | 1395 | 1695 | 19500 | 24 | 935.8 | GDDR6X | 384 | 134.4 162.72 | 459.20 555.96 | 29.38 35.68 | 0.459 0.558 | 29.38 35.68 | 235.08 285.48 | NaN | NaN | 69[125] | 350 | 2-way NVLink | NaN | $1499 |
pd[0].info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 5 entries, 0 to 4
Data columns (total 29 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 (Model, Model) 5 non-null object
1 (Launch, Launch) 5 non-null object
2 (Code name, Code name) 5 non-null object
3 (Process, Process) 5 non-null object
4 (Transistors (billion), Transistors (billion)) 5 non-null float64
5 (Die size (mm2), Die size (mm2)) 5 non-null float64
6 (Core config[a], Core config[a]) 5 non-null object
7 (Bus interface, Bus interface) 5 non-null object
8 (L2 Cache(MB), L2 Cache(MB)) 5 non-null int64
9 (Clock speeds, Base core clock (MHz)) 5 non-null int64
10 (Clock speeds, Boost core clock (MHz)) 5 non-null int64
11 (Clock speeds, Memory (MT/s)) 5 non-null int64
12 (Memory, Size (GiB)) 5 non-null int64
13 (Memory, Bandwidth (GB/s)) 5 non-null float64
14 (Memory, Bus type) 5 non-null object
15 (Memory, Bus width (bit)) 5 non-null int64
16 (Fillrate, Pixel (GP/s)) 3 non-null object
17 (Fillrate, Texture (GT/s)) 3 non-null object
18 (Processing power (TFLOPS), Single precision) 5 non-null object
19 (Processing power (TFLOPS), Double precision) 5 non-null object
20 (Processing power (TFLOPS), Half precision) 5 non-null object
21 (Processing power (TFLOPS), Tensor compute (FP16) (2:1 sparse)) 4 non-null object
22 (Ray-tracing Performance, Rays/s (Billions)) 0 non-null float64
23 (Ray-tracing Performance, RTX OPS/s (Trillions)) 0 non-null float64
24 (Ray-tracing Performance, Ray Perf TFLOPS) 4 non-null object
25 (TDP (Watts), TDP (Watts)) 5 non-null int64
26 (NVLink support, NVLink support) 5 non-null object
27 (Release price (USD), MSRP) 0 non-null float64
28 (Release price (USD), Founders Edition) 5 non-null object
dtypes: float64(6), int64(7), object(16)
memory usage: 1.3+ KBNotice that we have two levels of column names. This is because the original table has two headings for the column names. We don’t really need it, even though panda data frames can handle it. Let’s just keep the last level
df = pd[0]
df.columns = df.columns.get_level_values(1)
df.info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 5 entries, 0 to 4
Data columns (total 29 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Model 5 non-null object
1 Launch 5 non-null object
2 Code name 5 non-null object
3 Process 5 non-null object
4 Transistors (billion) 5 non-null float64
5 Die size (mm2) 5 non-null float64
6 Core config[a] 5 non-null object
7 Bus interface 5 non-null object
8 L2 Cache(MB) 5 non-null int64
9 Base core clock (MHz) 5 non-null int64
10 Boost core clock (MHz) 5 non-null int64
11 Memory (MT/s) 5 non-null int64
12 Size (GiB) 5 non-null int64
13 Bandwidth (GB/s) 5 non-null float64
14 Bus type 5 non-null object
15 Bus width (bit) 5 non-null int64
16 Pixel (GP/s) 3 non-null object
17 Texture (GT/s) 3 non-null object
18 Single precision 5 non-null object
19 Double precision 5 non-null object
20 Half precision 5 non-null object
21 Tensor compute (FP16) (2:1 sparse) 4 non-null object
22 Rays/s (Billions) 0 non-null float64
23 RTX OPS/s (Trillions) 0 non-null float64
24 Ray Perf TFLOPS 4 non-null object
25 TDP (Watts) 5 non-null int64
26 NVLink support 5 non-null object
27 MSRP 0 non-null float64
28 Founders Edition 5 non-null object
dtypes: float64(6), int64(7), object(16)
memory usage: 1.3+ KBNow it is a lot easier to use the column names as we only have one level of columns.
df| Model | Launch | Code name | Process | Transistors (billion) | Die size (mm2) | Core config[a] | Bus interface | L2 Cache(MB) | Base core clock (MHz) | Boost core clock (MHz) | Memory (MT/s) | Size (GiB) | Bandwidth (GB/s) | Bus type | Bus width (bit) | Pixel (GP/s) | Texture (GT/s) | Single precision | Double precision | Half precision | Tensor compute (FP16) (2:1 sparse) | Rays/s (Billions) | RTX OPS/s (Trillions) | Ray Perf TFLOPS | TDP (Watts) | NVLink support | MSRP | Founders Edition | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | GeForce RTX 3060[121] | February 25, 2021 | GA106-300-A1 | Samsung8N(10LP++++) | 13.25 | 300.0 | 3584:112:64:112:28(28) (?) | PCIe 4.0 x16 | 3 | 1320 | 1777 | 15000 | 12 | 360.0 | GDDR6 | 192 | NaN | NaN | 9.46 12.74 | 0.148 0.199 | 9.46 12.74 | NaN | NaN | NaN | 25 | 170 | No | NaN | $329 |
| 1 | GeForce RTX 3060 Ti[122] | December 2, 2020 | GA104-200-A1 | Samsung8N(10LP++++) | 17.40 | 392.5 | 4864:152:80:152:38(38) (6) | PCIe 4.0 x16 | 4 | 1410 | 1665 | 14000 | 8 | 448.0 | GDDR6 | 256 | NaN | NaN | 13.7 16.20 | 0.214 0.253 | 13.7 16.20 | ? 129.6 | NaN | NaN | NaN | 200 | No | NaN | $399 |
| 2 | GeForce RTX 3070[123] | October 29, 2020[124] | GA104-300-A1 | Samsung8N(10LP++++) | 17.40 | 392.5 | 5888:184:96:184:46(46) (6) | PCIe 4.0 x16 | 4 | 1500 | 1725 | 14000 | 8 | 448.0 | GDDR6 | 256 | 96.0 110.72 | 276.0 318.32 | 17.66 20.37 | 0.276 0.318 | 17.66 20.37 | 141.31 162.98 | NaN | NaN | 40[125] | 220 | No | NaN | $499 |
| 3 | GeForce RTX 3080[126] | September 17, 2020 | GA102-200-KD-A1 | Samsung8N(10LP++++) | 28.30 | 628.4 | 8704:272:96:272:68(68) (7) | PCIe 4.0 x16 | 5 | 1440 | 1710 | 19000 | 10 | 760.0 | GDDR6X | 320 | 126.72 150.48 | 391.68 465.12 | 25.06 29.76 | 0.392 0.465 | 25.06 29.76 | 200.54 238.14 | NaN | NaN | 58[125] | 320 | No | NaN | $699 |
| 4 | GeForce RTX 3090[127] | September 24, 2020 | GA102-300-A1 | Samsung8N(10LP++++) | 28.30 | 628.4 | 10496:328:112:328:82(82) (7) | PCIe 4.0 x16 | 6 | 1395 | 1695 | 19500 | 24 | 935.8 | GDDR6X | 384 | 134.4 162.72 | 459.20 555.96 | 29.38 35.68 | 0.459 0.558 | 29.38 35.68 | 235.08 285.48 | NaN | NaN | 69[125] | 350 | 2-way NVLink | NaN | $1499 |
Perfect! We got the table we wanted. Time to do some data clenup. Notice that the “Launch date” has an unwanted string[24] at the end. Let’s remove it
df["Launch"]0 February 25, 2021
1 December 2, 2020
2 October 29, 2020[124]
3 September 17, 2020
4 September 24, 2020
Name: Launch, dtype: objectLet’s get rid of it using a regular expression.
df['Launch'] = df['Launch'].str.replace(r"[d*?]","")df| Model | Launch | Code name | Process | Transistors (billion) | Die size (mm2) | Core config[a] | Bus interface | L2 Cache(MB) | Base core clock (MHz) | Boost core clock (MHz) | Memory (MT/s) | Size (GiB) | Bandwidth (GB/s) | Bus type | Bus width (bit) | Pixel (GP/s) | Texture (GT/s) | Single precision | Double precision | Half precision | Tensor compute (FP16) (2:1 sparse) | Rays/s (Billions) | RTX OPS/s (Trillions) | Ray Perf TFLOPS | TDP (Watts) | NVLink support | MSRP | Founders Edition | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | GeForce RTX 3060[121] | February 25, 2021 | GA106-300-A1 | Samsung8N(10LP++++) | 13.25 | 300.0 | 3584:112:64:112:28(28) (?) | PCIe 4.0 x16 | 3 | 1320 | 1777 | 15000 | 12 | 360.0 | GDDR6 | 192 | NaN | NaN | 9.46 12.74 | 0.148 0.199 | 9.46 12.74 | NaN | NaN | NaN | 25 | 170 | No | NaN | $329 |
| 1 | GeForce RTX 3060 Ti[122] | December 2, 2020 | GA104-200-A1 | Samsung8N(10LP++++) | 17.40 | 392.5 | 4864:152:80:152:38(38) (6) | PCIe 4.0 x16 | 4 | 1410 | 1665 | 14000 | 8 | 448.0 | GDDR6 | 256 | NaN | NaN | 13.7 16.20 | 0.214 0.253 | 13.7 16.20 | ? 129.6 | NaN | NaN | NaN | 200 | No | NaN | $399 |
| 2 | GeForce RTX 3070[123] | October 29, 2020 | GA104-300-A1 | Samsung8N(10LP++++) | 17.40 | 392.5 | 5888:184:96:184:46(46) (6) | PCIe 4.0 x16 | 4 | 1500 | 1725 | 14000 | 8 | 448.0 | GDDR6 | 256 | 96.0 110.72 | 276.0 318.32 | 17.66 20.37 | 0.276 0.318 | 17.66 20.37 | 141.31 162.98 | NaN | NaN | 40[125] | 220 | No | NaN | $499 |
| 3 | GeForce RTX 3080[126] | September 17, 2020 | GA102-200-KD-A1 | Samsung8N(10LP++++) | 28.30 | 628.4 | 8704:272:96:272:68(68) (7) | PCIe 4.0 x16 | 5 | 1440 | 1710 | 19000 | 10 | 760.0 | GDDR6X | 320 | 126.72 150.48 | 391.68 465.12 | 25.06 29.76 | 0.392 0.465 | 25.06 29.76 | 200.54 238.14 | NaN | NaN | 58[125] | 320 | No | NaN | $699 |
| 4 | GeForce RTX 3090[127] | September 24, 2020 | GA102-300-A1 | Samsung8N(10LP++++) | 28.30 | 628.4 | 10496:328:112:328:82(82) (7) | PCIe 4.0 x16 | 6 | 1395 | 1695 | 19500 | 24 | 935.8 | GDDR6X | 384 | 134.4 162.72 | 459.20 555.96 | 29.38 35.68 | 0.459 0.558 | 29.38 35.68 | 235.08 285.48 | NaN | NaN | 69[125] | 350 | 2-way NVLink | NaN | $1499 |
Lets convert the launch date to a timestamp now
df['Launch'] = pandas.to_datetime(df['Launch'])
df.info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 5 entries, 0 to 4
Data columns (total 29 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 Model 5 non-null object
1 Launch 5 non-null datetime64[ns]
2 Code name 5 non-null object
3 Process 5 non-null object
4 Transistors (billion) 5 non-null float64
5 Die size (mm2) 5 non-null float64
6 Core config[a] 5 non-null object
7 Bus interface 5 non-null object
8 L2 Cache(MB) 5 non-null int64
9 Base core clock (MHz) 5 non-null int64
10 Boost core clock (MHz) 5 non-null int64
11 Memory (MT/s) 5 non-null int64
12 Size (GiB) 5 non-null int64
13 Bandwidth (GB/s) 5 non-null float64
14 Bus type 5 non-null object
15 Bus width (bit) 5 non-null int64
16 Pixel (GP/s) 3 non-null object
17 Texture (GT/s) 3 non-null object
18 Single precision 5 non-null object
19 Double precision 5 non-null object
20 Half precision 5 non-null object
21 Tensor compute (FP16) (2:1 sparse) 4 non-null object
22 Rays/s (Billions) 0 non-null float64
23 RTX OPS/s (Trillions) 0 non-null float64
24 Ray Perf TFLOPS 4 non-null object
25 TDP (Watts) 5 non-null int64
26 NVLink support 5 non-null object
27 MSRP 0 non-null float64
28 Founders Edition 5 non-null object
dtypes: datetime64[ns](1), float64(6), int64(7), object(15)
memory usage: 1.3+ KBHaving a valid datetime column will help with plotting graphs.
#October 29, 2020[124]
def customDateParser(x):
dateStr = x.replace("[24]","");
return pd.datetime.strptime(dateStr, "%M %d, %Y")
pd = pandas.read_html("https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_processing_units", match="10496:328:112:328:82")pd[0].info()<class 'pandas.core.frame.DataFrame'>
RangeIndex: 5 entries, 0 to 4
Data columns (total 29 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 (Model, Model) 5 non-null object
1 (Launch, Launch) 5 non-null object
2 (Code name, Code name) 5 non-null object
3 (Process, Process) 5 non-null object
4 (Transistors (billion), Transistors (billion)) 5 non-null float64
5 (Die size (mm2), Die size (mm2)) 5 non-null float64
6 (Core config[a], Core config[a]) 5 non-null object
7 (Bus interface, Bus interface) 5 non-null object
8 (L2 Cache(MB), L2 Cache(MB)) 5 non-null int64
9 (Clock speeds, Base core clock (MHz)) 5 non-null int64
10 (Clock speeds, Boost core clock (MHz)) 5 non-null int64
11 (Clock speeds, Memory (MT/s)) 5 non-null int64
12 (Memory, Size (GiB)) 5 non-null int64
13 (Memory, Bandwidth (GB/s)) 5 non-null float64
14 (Memory, Bus type) 5 non-null object
15 (Memory, Bus width (bit)) 5 non-null int64
16 (Fillrate, Pixel (GP/s)) 3 non-null object
17 (Fillrate, Texture (GT/s)) 3 non-null object
18 (Processing power (TFLOPS), Single precision) 5 non-null object
19 (Processing power (TFLOPS), Double precision) 5 non-null object
20 (Processing power (TFLOPS), Half precision) 5 non-null object
21 (Processing power (TFLOPS), Tensor compute (FP16) (2:1 sparse)) 4 non-null object
22 (Ray-tracing Performance, Rays/s (Billions)) 0 non-null float64
23 (Ray-tracing Performance, RTX OPS/s (Trillions)) 0 non-null float64
24 (Ray-tracing Performance, Ray Perf TFLOPS) 4 non-null object
25 (TDP (Watts), TDP (Watts)) 5 non-null int64
26 (NVLink support, NVLink support) 5 non-null object
27 (Release price (USD), MSRP) 0 non-null float64
28 (Release price (USD), Founders Edition) 5 non-null object
dtypes: float64(6), int64(7), object(16)
memory usage: 1.3+ KBpd[0]["Launch"]| Launch | |
|---|---|
| 0 | February 25, 2021 |
| 1 | December 2, 2020 |
| 2 | October 29, 2020[124] |
| 3 | September 17, 2020 |
| 4 | September 24, 2020 |
