Nvidia GeForce 100 series

GeForce GeForce 100 series is a series of desktop - graphics chip company Nvidia , which only in the OEM is available -Markt.

GeForce series arose from the marketing need to classify the existing graphics chips of the GeForce 9 series in the new naming scheme of the GeForce 200 series, which was published shortly before . Despite its name, the GeForce 100 series can therefore be classified chronologically after the GeForce 200 series, but technically identical to the GeForce 9 series.

Technical

GeForce chips of the GeForce 100 series are identical to the respective chips of the GeForce 9 series and therefore all support the Shader model 4.0 (SM 4.0) according to DirectX 10, OpenGL 3.3 , CUDA and thus also PhysX .

Graphics processors

Graphics chip	production			units						L2 cache	API support			Video pro- cessor	Bus interface stelle
	production process	transis- interfere	The surface	ROP particle functions	ROPs	Unified shaders		Texture units			DirectX	OpenGL	OpenCL
						Stream processors	Shader - cluster	TAUs	TMUs
G92b	55 nm	754 million	276 mm²	4th	16	128	8th	64	64	k. A.	10.0	3.3	1.1	VP2	PCIe 2.0
G94b		505 million	180 mm²	4th	16	064	4th	32	32
G96b		314 million	119 mm²	2	08th	032	2	16	16
G98	65 nm	210 million	086 mm²	1	04th	008th	1	08th	08th					VP3

Naming

The GeForce 100 series uses a naming scheme similar to that used since the GeForce 200 series . All graphics chips are identified with a letter abbreviation to classify the performance sector and a three-digit number that generally begins with a “1” (for GeForce 100). The last two digits serve for further differentiation within the respective service sector. Since the GeForce 200 series, which appeared earlier, already covered the high-end market, the abbreviation GTX was not used in the 100 series.

Letter abbreviations

• G - Low-Budget
• GT - mainstream
• GTS - performance
• GTX - high-end

Confusion is caused by the fact that Nvidia already had and has the GeForce 100 cards on the market as GeForce 9 cards (as of January 2009), which in turn already existed technically identically in the GeForce 8 series. Corresponding renaming was carried out, first to the naming scheme of the 9 series and now to the scheme of the 100 series (e.g. 9500 GT → GT 120). This means that one and the same cards are sometimes on the market under three different names.

Model data

model	Official launch	Graphics processor (GPU)							Graphics memory				MGCP ( watt )	Others
		Type	Active units				Chip clock (MHz)	Shader clock (MHz)	Size ( MB )	Clock rate (MHz)	Type	Storage interface
			ROPs	Shader - cluster	Stream processors	Texture units
GeForce G100	March 10, 2009	G98	4th	1	8th	8th	567	1400	512	500	DDR2	64 bit	35	OEM version of the GeForce 9300 GS
GeForce GT 120	March 10, 2009	G96b	8th	2	32	16	500	1400	512	500	DDR2	128 bit	50	OEM version of the GeForce 9500 GT
GeForce GT 130	March 10, 2009	G94b	12	3	48	24	500	1250	768	800	GDDR3	192 bits	75	OEM version of the GeForce 9600 GSO 512
GeForce GTS 150	March 10, 2009	G92b	16	8th	128	64	738	1836	1024	1000	GDDR3	256 bit	141	OEM version of the GeForce 9800 GTX +

Hints

• GeForce specified clock rates are the recommended or specified by Nvidia. However, the final specification of the clock rates is in the hands of the respective graphics card manufacturer. It is therefore quite possible that there are or will be graphics card models that have different clock rates.
• The date indicated is the date of the public presentation, not the date of availability of the models.
• "MGCP" stands for "Maximum Graphics Card Power" and describes the maximum power consumption of a graphics card that is officially specified by nVidia and can be expected under normal operation. Under certain conditions, however, an even higher power consumption can be achieved. Therefore, the Maximum Graphics Card Power cannot be equated with the TDP .
• GeForce clock frequency of the memory is also often given as twice as high. The reason for this is the double data rate (DDR).

Performance data

The following theoretical performance data result for the respective models:

model	Computing power of all stream processors ( GFlops )	Fill rate of the graphics processor		Data transfer rate to graphics memory ( GB / s)
		Pixel (GPixel / s)	Texel ( GTexel / s)
GeForce G100	33.6	2.3	4.5	8.0
GeForce GT 120	134	4.0	8.0	16.0
GeForce GT 130	180	6.0	12.0	38.4
GeForce GTS 150	705	11.8	47.2	64.0

Hints

• GeForce specified performance values ​​for the computing power via the stream processors, the pixel fill rate, the texel fill rate and the memory bandwidth are theoretical maximum values. GeForce overall performance of a graphics card depends, among other things, on how well the available resources can be used or utilized. There are also other factors that are not listed here that affect performance.
• GeForce specified computing power via the stream processors refers to the use of both MUL operations, which is not achieved with graphics shader calculations, since further calculations have to be carried out. In these calculations, the computing power of the stream processors is therefore lower.

Web links

• Product overview on the manufacturer's website

Individual evidence

1. ↑ ComputerBase: Two entry-level cards from Nvidia in 40 nm? , Message dated January 28, 2009, accessed February 1, 2010
2. ↑ http://www.nvidia.de/object/win7-winvista-32bit-257.21-whql-driver-de.html OpenCL 1.0 from Tesla chip G80 with WHQL 257.21
3. ↑ http://www.gpu-tech.org/content.php/162-Nvidia-supports-OpenCL-1.1-with-GeForce-280.19-Beta-performance-suffers OpenCL 1.1 from Tesla chip G80 with Beta 280.19
4. ↑ Product page: NVIDIA GeForce G100 , accessed February 1, 2010
5. ↑ Product page: NVIDIA GeForce GT 120 , accessed February 1, 2010
6. ↑ Product page: NVIDIA GeForce GT 130 , accessed February 1, 2010
7. ↑ Product page: NVIDIA GeForce GTS 150 , accessed February 1, 2010