GeForce GTS 450 Video Card Review
By
Gabriel Torres
on September 24, 2010
The GeForce GTS 450 is the first DirectX 11 mid-range graphics chip released by NVIDIA, and today we are publishing the review of the NVIDIA reference model. Let’s see if video cards based on this new chip are worthwhile buying.
The new GTS 450 comes on the same price range as the GeForce GTS 250 with 1 GB (which is a DirectX 10 part), making it to compete directly with the Radeon HD 5750 and indirectly with the Radeon HD 5770 (as it is a little bit more expensive), which are on the market for almost a year now.
In the table below we list all the main technical specification of the video cards included in our comparison.
Video Card | Core Clock | Shader Clock | Memory Clock (Real) | Memory Clock (Effective) | Memory Interface | Memory Transfer Rate | Memory | Shaders | DirectX |
GeForce GTS 250 | 738 MHz | 1,836 MHz | 1.1 GHz | 2.2 GHz | 256-bit | 70.4 GB/s | 1 GB GDDR3 | 128 | 10 |
GeForce GTS 450 | 783 MHz | 1,566 MHz | 902 MHz | 3,608 MHz | 128-bit | 57.7 GB/s | 1 GB GDDR5 | 192 | 11 |
Radeon HD 5750 | 700 MHz | 700 MHz | 1.15 GHz | 4.6 GHz | 128-bit | 73.6 GB/s | 1 GB GDDR5 | 720 | 11 |
Radeon HD 5770 | 850 MHz | 850 MHz | 1.2 GHz | 4.8 GHz | 128-bit | 76.8 GB/s | 1 GB GDDR5 | 800 | 11 |
You can compare the specs of these video cards with other video cards by taking a look at our AMD ATI Chips Comparison Table and NVIDIA Chips Comparison Table tutorials.
Now let’s take an in-depth look at the GeForce GTS 450 reference sample from NVIDIA.
The video card we are reviewing is the NVIDIA reference model, which is usually the exact same model video card manufacturers sell, only adding their sticker to it. So, the performance of GeForce GTS 250 video cards from NVIDIA partners should be exactly the same as the one achieved by the reviewed sample – unless, of course, the NVIDIA partner adds some overclocking to the video card. In order to differentiate themselves on the market, some manufacturers may also choose to add their own cooling solution.
click to enlarge
Figure 1: GeForce GTS 450
click to enlarge
Figure 2: GeForce GTS 450
This video card has three video outputs, two DVI-D and one HDMI.
click to enlarge
Figure 3: Video connectors
In Figure 4, you can see the video card with its cooler removed. Note how it requires one six-pin auxiliary power connector. In Figures 5 and 6, you can see the cooler by itself. The cooler used on the GeForce GTS 450 is made of aluminum and cools down only the GPU.
click to enlarge
Figure 4: Video card with the cooler removed
click to enlarge
Figure 5: The GPU cooler
click to enlarge
Figure 6: The GPU cooler
The reviewed card uses eight 1 Gbit GDDR5 chips (four soldered on each side of the printed circuit board), making its 1 GB video memory (1 Gbit x 8 = 1 GB).
The chips used are K4G10325FE-HC05 parts from Samsung, which support up to 1 GHz (4 GHz QDR) and since on this video card memory is accessed at 902 MHz (3,608 MHz QDR), there is a good margin for you to increase the memory clock rate while keeping the chips inside the maximum they support. Of course you can always try to overclock the memory chips above their specs.
click to enlarge
Figure 7: Memory chips
Before seeing the performance results, let’s recap the main features of this video card.
The main specifications for the GeForce GTS 450 video card include:
* Researched at Newegg.com on the day we published this review.
During our benchmarking sessions, we used the configuration listed below. Between our benchmarking sessions the only variable was the video card being tested.
Hardware Configuration
Software Configuration
Driver Versions
Software Used
Error Margin
We adopted a 3% error margin. Thus, differences below 3% cannot be considered relevant. In other words, products with a performance difference below 3% should be considered as having similar performance.
3DMark Vantage measures Shader 4.0 (i.e., DirectX 10) performance and supports PhysX, a programming interface developed by Ageia (now part of NVIDIA) to transfer physics calculations from the system CPU to the video card GPU in order to increase performance. Mechanical physics is the basis for calculations about the interaction of objects. For example, if you shoot, what exactly will happen to the object when the bullet hits it? Will it break? Will it move? Will the bullet bounce back? Note that since we are considering only the GPU score provided by this program, physics calculations are not taken into account.
We ran this program at two 16:10 widescreen resolutions, 1680x1050 and 1920x1200. First we used the “Performance” profile, and then we used the “Extreme” profile (basically enabling anti-aliasing at 4x, anisotropic filtering at 16x, and putting all detail settings at their maximum or “extreme” values). The results being compared are the “GPU Scores” achieved by each video card.
3DMark Vantage – Performance | 1680x1050 | Difference |
Radeon HD 5770 | 7578 | 15.3% |
GeForce GTS 450 | 6572 |
|
Radeon HD 5750 | 6098 | -7.2% |
GeForce GTS 250 | 5143 | -21.7% |
| 3DMark Vantage – Performance | 1920x1200 | Difference |
Radeon HD 5770 | 5932 | 16.0% |
GeForce GTS 450 | 5112 |
|
Radeon HD 5750 | 4808 | -5.9% |
GeForce GTS 250 | 3908 | -23.6% |
3DMark Vantage – Extreme | 1680x1050 | Difference |
Radeon HD 5770 | 5613 | 4.0% |
GeForce GTS 450 | 5399 |
|
Radeon HD 5750 | 4492 | -16.8% |
GeForce GTS 250 | 3924 | -27.3% |
3DMark Vantage – Extreme | 1920x1200 | Difference |
Radeon HD 5770 | 4502 | 6.4% |
GeForce GTS 450 | 4231 |
|
Radeon HD 5750 | 3582 | -15.3% |
GeForce GTS 250 | 3022 | -28.6% |
StarCraft II: Wings of Liberty is a very popular DirectX 9 game released earlier this year. Though this game uses an old version of DirectX, the number of textures that can be represented on one screen can push most of the top-end graphics cards to their limits (especially when the graphics settings are set at “Ultra”). StarCraft II: Wings of Liberty uses its own physics engine that is bound to the CPU and thus does not benefit from PhysX.
As this game is a very competitive and widespread “e-sport,” many LAN cafés and competitions will be building systems to run this game at high frame rates, but at lower graphics settings. For gamers that play StarCraft II: Wings of Liberty at a higher level, “Ultra” graphics settings can be a hindrance due to clutter. Due to this, the more popular video cards for this game will be the mid-range cards.
We tested this game on two 16:10 resolution settings, 1680x1050 and 1920x1200. The quality of the game was set to the medium preset and the textures were also set to medium. We used the NVIDIA and ATI consoles to force both Anti-Aliasing and Anisotropic Filtering at 4x. We then used FRAPS to collect the frame rate of a replay on the “Unit Testing” custom map. We used a battle between very large armies to stress the video cards.
Starcraft II – Medium | 1680x1050 | Difference |
GeForce GTS 450 | 102.46 |
|
Radeon HD 5770 | 96.76 | -5.6% |
Radeon HD 5750 | 90.26 | -11.9% |
GeForce GTS 250 | 82.68 | -19.3% |
Starcraft II – Medium | 1920x1200 | Difference |
GeForce GTS 450 | 92.52 |
|
Radeon HD 5770 | 90.24 | -2.5% |
Radeon HD 5750 | 84.67 | -8.5% |
GeForce GTS 250 | 64.30 | -30.5% |
Call of Duty 4 is a DirectX 9 game implementing high-dynamic range (HDR) and its own physics engine, which is used to calculate how objects interact. For example, if you shoot, exactly what will happen to the object when the bullet hits it? Will it break? Will it move? Will the bullet bounce back? It gives a more realistic experience to the user.
To get accurate results we had to disable the 80 FPS limit in the game. To do this, input the command, “/seta com_maxfps 1000” (minus the quotes) into the console (` key). It can be set to any number greater than 200.
We ran this program at two 16:10 widescreen resolutions, 1680x1050 and 1920x1200, maxing out all image quality controls (i.e., everything was set to the maximum values in the Graphics and Texture menus). We used the internal game benchmarking feature, running a demo provided by NVIDIA called “wetwork.” We are putting this demo here for downloading if you want to run your own benchmarks. We ran the demo five times, and the results below are the average number of frames per second (FPS) achieved by each video card.
Call of Duty 4 – Maximum | 1680x1050 | Difference |
GeForce GTS 250 | 90.2 | 1.81% |
GeForce GTS 450 | 88.6 |
|
Radeon HD 5770 | 83.9 | -5.3% |
Radeon HD 5750 | 75.5 | -14.8% |
Call of Duty 4 – Maximum | 1920x1200 | Difference |
GeForce GTS 250 | 71.7 | 8.92% |
Radeon HD 5770 | 71.5 | 8.59% |
GeForce GTS 450 | 65.8 |
|
Radeon HD 5750 | 63.2 | -4.0% |
Crysis Warhead is a DirectX 10 game based on the same engine as the original Crysis, but optimized (it runs under DirectX 9.0c when installed on Windows XP).
We used the HardwareOC Crysis Warhead Benchmark Tool to collect the data for this test. We ran this program at two 16:10 widescreen resolutions, 1680x1050 and 1920x1200, all at low image quality (no anti-aliasing and no anisotropic filtering) and using the Airfield demo. The results below are the number of frames per second achieved by each video card.
Crysis Warhead – Medium | 1680x1050 | Difference |
GeForce GTS 450 | 42 |
|
GeForce GTS 250 | 41 | -2.4% |
Radeon HD 5770 | 41 | -2.4% |
Radeon HD 5750 | 36 | -14.3% |
Crysis Warhead – Medium | 1920x1200 | Difference |
Radeon HD 5770 | 37 | 2.78% |
GeForce GTS 450 | 36 |
|
GeForce GTS 250 | 36 | 0.0% |
Radeon HD 5750 | 32 | -11.1% |
Far Cry 2 is based on an entirely new game engine called Dunia, which is DirectX 10 when played under Windows Vista with a DirectX 10 compatible video card.
We used the benchmarking utility that comes with this game, setting image quality to High (x4 anti-aliasing and no anisotropic filtering) and running the “Ranch Long” demo three times. The results below are expressed in frames per second and are an arithmetic average of the three results collected.
FarCry 2 – Maximum | 1680x1050 | Difference |
GeForce GTS 250 | 82.7 | 25.8% |
GeForce GTS 450 | 65.7 |
|
Radeon HD 5770 | 60.7 | -7.7% |
Radeon HD 5750 | 52.4 | -20.3% |
FarCry 2 – Maximum | 1920x1200 | Difference |
GeForce GTS 250 | 64.3 | 20.3% |
GeForce GTS 450 | 53.5 |
|
Radeon HD 5770 | 53.3 | -0.4% |
Radeon HD 5750 | 46.3 | -13.5% |
Aliens vs. Predator is a DirectX 11 game that makes full use of tessellation and advanced shadow rendering. We used the Aliens vs. Predator Benchmark Tool developed by Rebellion. This program reads its configuration from a text file (our configuration files can be found here). We ran this program at 1680x1050 and 1920x1200 resolutions, with low texture settings, x4 anisotropic filtering and x4 anti-aliasing.
Aliens vs. Predator | 1680x1050 | Difference |
Radeon HD 5770 | 29.9 | 2.0% |
GeForce GTS 450 | 29.3 |
|
Radeon HD 5750 | 27.6 | -5.8% |
GeForce GTS 250 | 0 | - |
Aliens vs. Predator | 1920x1200 | Difference |
Radeon HD 5770 | 28.5 | 8.8% |
Radeon HD 5750 | 26.5 | 1.1% |
GeForce GTS 450 | 26.2 |
|
GeForce GTS 250 | 0 | - |
Metro 2033 is a DirectX 11 game. To benchmark this game we used FRAPS to record the average FPS while playing part way through the Chase level (the save file we used can be downloaded here). We ran this program at 1680x1050 and 1920x1200 resolutions, setting texture quality at “Low,” anti-aliasing at “AAA,” anisotropic filtering at 4x, tessellation “on,” and DoF “off.”
Metro 2033 | 1680x1050 | Difference |
Radeon HD 5770 | 54.76 | 13.8% |
Radeon HD 5750 | 50.28 | 4.5% |
GeForce GTS 450 | 48.10 |
|
GeForce GTS 250 (DX10) | 47.40 | -1.5% |
Metro 2033 | 1920x1200 | Difference |
Radeon HD 5770 | 46.64 | 20.5% |
Radeon HD 5750 | 42.69 | 10.3% |
GeForce GTS 250 (DX10) | 39.47 | 2.0% |
GeForce GTS 450 | 38.71 |
Darkest of days is a DirectX 9 game that implements a PhysX engine, moving physics calculations from the CPU to the GPU. Although it’s not very popular, we added this game because of its PhysX benchmarking feature. We ran this game at 1680x1050 with details set at “very high,” and both anti-aliasing and anisotropic filtering disabled. We ran three tests, first with PhysX set at “low,” where the game makes all physics calculations using the system CPU, then increasing it to “medium” (which adds leaves, wind and weapons impact effects due to bullets and grenades), and finally increasing it to “high” (which adds fog and smoke effects). The medium and high PhysX levels move physics calculations from the CPU to the GPU. Keep in mind that only NVIDIA-based cards support PhysX.
Darkest of Days - 1680x1050 | Low PhysX | Difference |
GeForce GTS 450 | 112.63 |
|
Radeon HD 5770 | 108.65 | -3.5% |
Radeon HD 5750 | 105.68 | -6.2% |
GeForce GTS 250 | 73.10 | -35.1% |
Darkest of Days - 1680x1050 | Medium PhysX | Difference |
GeForce GTS 450 | 76.48 |
|
GeForce GTS 250 | 50.06 | -34.5% |
Radeon HD 5750 | 2.91 | -96.2% |
Radeon HD 5770 | 2.70 | -96.5% |
Darkest of Days - 1680x1050 | High PhysX | Difference |
GeForce GTS 450 | 42.46 |
|
GeForce GTS 250 | 29.60 | -30.3% |
Radeon HD 5750 | 1.42 | -96.6% |
Radeon HD 5770 | 1.28 | -97.0% |
We were truly impressed by the new GeForce GTS 450. From the eight games and simulations we ran, the GTS 450 only lost to its main competitor, the Radeon HD 5750, in one (Metro 2033). In all others it was either faster or achieved the same performance level.
The improvement in PhysX performance compared to the previous generation mid-range video card from NVIDIA, the GeForce GTS 250, was also impressive (43% improvement as measured by Darkest of Days).
If you are looking for a mid-range card in the USD 130 price range, the new GeForce GTS 450 is the way to go.
Originally at http://www.hardwaresecrets.com/article/GeForce-GTS-450-Video-Card-Review/1095