discussion ASUS GeForce GT 430 1GB Review

[SKYMTL]

DiRT 2 (DX11)

DiRT 2 (DX11)

Being one of the newest games on the market, DiRT 2 cuts an imposing figure in terms of image quality and effects fidelity. We find that to benchmark this game the in-game tool is by far the best option. However, due to small variances from one race to another, three benchmark runs are done instead of the normal two. It should also be mentioned that the demo version of the game was NOT used since after careful testing, the performance of the demo is not representative of the final product. DX11 was forced through the game’s config file. In addition, you will see that these scores do not line up with our older benchmarks at all. This is due to the fact that a patch was recently rolled out for the game which included performance optimizations in addition to new graphics options.


1280 x 800

1440 x 900

The GT 430’s performance in Dirt 2 is literally identical to the 8 month old HD 5570 1GB with its only partially convincing win coming with the minimum framerates with AA enabled at 1440 x 900.

 

[SKYMTL]

StarCraft II (DX9)

StarCraft II (DX9)

In order to test StarCraft II performance we recorded a typical multiplayer battle on the Agria Valley map and used it as a replay in order to ensure every run was identical to one another. We used the last 3 minutes of the replay which includes the final assault on the enemy base. MSAA was applied in the NVIDIA and ATI control panels for certain tests.

1440 x 900

StarCraft II once again shows the GT 430 neck and neck with the HD 5570 but what impressed us was how much of an improvement it showed over the outgoing GT 220.

 

[SKYMTL]

The GT 430 as a Dedicated PhysX Card

The GT 430 as a Dedicated PhysX Card

Mafia II uses some of the most intense physics calculations seen in any game through the use of PhysX’s APEX modules. For this test we used a typical sequence in Chapter 5 which can be seen below in order to benchmark PhysX performance in a number of configurations.

<iframe src="http://player.vimeo.com/video/14377977?byline=0&portrait=0" width="640" height="360" frameborder="0"></iframe>

Even though the rendering horsepower of the GT 430 is light years ahead of the GT 220, the amount of additional performance you get from using it as a dedicated PhysX card is negligible at best. There is a slight improvement but if you are already using a lower-end card for PhysX processing, you won’t see a noticeable benefit from “moving up” to a GT 430.

 

[SKYMTL]

HD Video & Audio Playback (Blu Ray)

HD Video & Audio Playback (Blu Ray)

One of the main selling points of any budget friendly graphics card is its supposed ability to decode the high definition images that define current generation Blu Rays.

In this test we take two demanding movies (Avatar and The Dark Knight) which have each been encoded in a different format and play them back in order to determine CPU usage. The signals are passed through to an Onkyo SR-507 receiver which has the ability to detect and play back lossless audio as well. For both clips, lossless audio is enabled through PowerDVD 10 Ultra.

There is however one thing to note: the peak CPU usage should be taken with a grain of salt since and is included for reference purposes only. Even though every other unnecessary process has been disabled, there will always be some CPU usage not being dedicated to video playback. This means “peaks” could very well be influenced by sources outside a GPU’s ability to play back HD content.


Avatar (MPEG4 AVC + DTS HD Master Audio)

NVIDIA’s Purevideo HD decoder had absolutely no issue with what many consider to be one of the most demanding Blu Rays currently on the market. It was able to keep average and maximum CPU usage to a minimum but then again, the difference between it and the other cards here is minimal at best and you won’t feel a performance difference between them when playing a movie.


The Dark Knight (VC-1 + Dolby TrueHD)

Much like Avatar, playback of movies sporting the VC-1 codec proved to be no problem for any of the cards we included. An average though, the NVIDIA products did do marginally better when it came to average CPU usage.


Lossless Audio Playback; Hit or Miss?

When the GTX 460 was first released, there were quite a few complaints on various AV forums that it flat out refused to support playback of lossless audio tracks. We experienced the same thing with the pre-release and beta drivers even though according to NVIDIA it was supposed to work. Now, months after the initial problems we decided to have another crack at it and....

...success! Both Dolby TrueHD and DTS MA tracks played back without an issue across every Blu Ray we tried on the GT 430.

There is however a number of things you should know before assuming your setup will automatically detect and play lossless audio tracks. To begin with, it is necessary to install NVIDIA’s HD Audio driver which is a selectable option in their new driver installation panel. Making sure your chosen media player (and receiver) support TrueHD and its ilk should go without mentioning but in order to ensure content bitstreaming, programs like PowerDVD have a setting to send non decoded high definition audio to an external device. Finally, most Blu Rays don’t have lossless audio playback selected by default so you’ll need to go into the on-screen menu and set it accordingly. After those hoops are jumped through, just sit back and enjoy the bliss of HD audio.

 

[SKYMTL]

Video Transcoding / Folding @ Home Performance

Video Transcoding / Folding @ Home Performance

Graphics Plus technologies have been championed by NVIDIA and AMD / ATI for the last few years as a means to speed up certain tasks which would normally take much longer to complete on the CPU. Naturally, media center PCs don’t always have the fastest possible processor so moving things like video and audio transcoding to the dedicated GPU only makes sense. In order to test performance we chose to transcode a HD video (complete with audio) and run some Folding @ Home.

HD Video Transcoding

Unfortunately, finding a video transcoding program which is customizable and supports CUDA is easy but finding one which is compatible with ATI’s Stream is next to impossible. We finally settled on Cyberlink’s MediaEspresso 6.

The file used is a 600MB 1080P .m2ts rip of Samsung’s absolutely stunning The Beauty of Nature demo disk which is then converted into a typical H.264 MP4 format that is used by all of Apple’s devices.

There should be no doubt in anyone’s mind that NVIDIA’s cards simply dominate this benchmark through an efficient use of CUDA. ATI’s cards on the other hand are slow at best and are beaten quite handily by the quad core Intel i5 750.

One odd thing that we noticed was the parity of the GT 220 and GT 430 in this particular test. Upon further investigation, it seems like not matter which NVIDA GPU was used the transcoding process was always performed at the same speed. GPU-Z showed us a mere 30% usage on the GT 430 while the GT 220 was running along at 55-60% which leads us to believe that the extra horsepower of the Fermi card is simply going to waste. Granted, it proves the GT 430 is much more efficient at transcoding but this doesn’t mean a darn thing if the task isn’t finished any quicker. Let’s hope that someday NVIDIA’s drivers will allow applications like this to take FULL advantage of higher end GPUs.

Folding @ Home Performance

One of the cornerstones of consumer GPGPU performance has always been a card’s ability to crunch numbers for distributed computing projects like Stanford’s Folding @ Home program.

In this test, we use the latest GPU3 client on a number of cards and let it run for 24 hours in order to get an approximate PPD average. Since this is the beta of GPU3, the quantity of projects is limited which helps give a slightly more accurate comparison between the GPUs. In order to monitor PPD, we used the HFM.net program. Note that Fermi card results are based off of a 611 point WU.

For a 49W, $80 graphics card the GT 430 throws down one hell of a gauntlet in our Folding chart. Its closest competitor is the GT 240 1GB which has the same number of shaders but has lower clock speeds while the GT 220 languishes in another dimension. One interesting thing to note is the near-linear scaling of performance in relation to the number of CUDA cores when the GT 430 is compared to the GTS 450.

Edit: Additional tests were carried out after the review was posted. On a 925 point GPU3 WU, the GT 430 achieved 8155 PPD.

 

[SKYMTL]

GPU Accelerated Internet Browsing

GPU Accelerated Internet Browsing

The days of mostly static webpages will soon be behind us as all the major browsers move towards the adoption of tools which can make the internet experience more immersive. Adding 3 dimensional graphics to websites and near-seamless zooms on Google / Bing Maps is only the tip of the iceberg. In order to provide users with this next generation experience, developers have begun taking advantage of the parallel processing power found in today’s GPUs. Through the use of hardware accelerated HTML5 and Direct2D (along with DirectWrite) we will soon start seeing some stunning additions to our favourite websites.

In order to test performance in these two areas, we used Microsoft’s Internet Explorer 9 Beta as well as their new IE9 Test Drive page. Note that if you want to replicate any of these tests you will need the IE9 beta as well as Windows Vista SP2 or Windows 7.


Direct2D Performance

To test Direct2D we loaded Microsoft’s Flying Images benchmark (which also features background compiled Javascript) and ensured the refresh rate of the Acer monitor was set to 120Hz in order to remove any 2D bottlenecks. The framerate was calculated by the in-benchmark counter and represents the average framerate after a 60 second demo with the maximum number of images on-screen.

If anything, this should drive home the need for a dedicated (or at least very good integrated) GPU. There really isn’t all that much difference between the NVIDIA and ATI sides of the fence but the GeForce cards were able to maintain a slim lead.

Much like in the video transcoding test, it seems like something is stopping higher end NVIDIA cards from pulling away from their forefathers.


GPU Accelerated HTML 5 Performance

Microsoft includes a perfect “worst case scenario” HTML 5 benchmark within their IE9 Test Drive page called Video Kaleidoscope. Not only does this play back a video but it also has the ability to add on-screen effects which can seriously tax your hardware. We used the Halo Reach demo along with all of the visual bangs and whistles enabled. Framerate logging was done through the FRAPS 2D benchmarking tool.

Like we said: this isn’t an easy test on any bit of hardware. The i5 750 CPU flatly refused to run it at all and even the entry level GPUs just couldn’t get it to run at the recommended 30 frames per second. Ironically, the biggest disappointment was actually the GT 430 since it didn’t even match the performance of a GT 220. All of the indicators point to a driver issue here but the results are sill nonetheless disappointing for the latest addition to the Fermi family.

 

[SKYMTL]

Adobe Flash 10.1 Acceleration

Adobe Flash 10.1 Acceleration

With YouTube’s move to high definition videos, many viewers suddenly found that they could not watch the newer, better looking videos without their computer getting bogged down. Granted, this is mostly an issue for netbooks and laptops but there are still plenty of lower-end and dual and single core desktop processors out there which just don’t have the horsepower necessary to process 720P or 1080P content. This means more often than not, that HD video you are intent on watching quickly turns into a slide show.

Adobe’s Flash 10 does offer some basic hardware acceleration for HD video streaming from YouTube, Hulu and other online sources but the 10.1 version takes things to the next level by offering Flash video processing which is accelerated by the GPU. This frees up CPU resources, allows for smoother playback on low-end systems and also provides an added benefit of (supposedly) higher picture quality.

In this test we used the Transformers 1080P trailer from Youtube while logging CPU usage in the Windows 7 Performance Monitor.

Other than the fact that the CPU shows what can happen without hardware acceleration enabled, all of the dedicated GPUs exhibit basically the same performance right across the board. Honestly, this test just doesn’t stress even low end GPUs all that much.

 

[SKYMTL]

Core Temperature & Acoustics

Core Temperature & Acoustics

For all temperature testing, the cards were placed on an open test bench with a single 120mm 1200RPM fan placed ~8” away from the heatsink. The ambient temperature was kept at a constant 22°C (+/- 0.5°C). If the ambient temperatures rose above 23°C at any time throughout the test, all benchmarking was stopped. For this test we use the 3DMark Batch Size test at it highest triangle count with 4xAA and 16xAF enabled and looped it for one hour to determine the peak load temperature as measured by GPU-Z.

For Idle tests, we let the system idle at the Vista desktop for 15 minutes and recorded the peak temperature.

Considering the size of this card’s heatsink, the temperatures aren’t all that bad but there were some issues with the way it achieves reasonable thermal performance. ASUS fitted their card with small fan that is relatively quiet at idle and minor loads but when gaming, things get a bit too loud for our liking. Luckily though, the light loads encountered when playing back a Blu Ray or watching a Youtube video don’t push fan speeds all that high so we can’t really complain all that much.

System Power Consumption

For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we once again use the Batch Render test in 3DMark06 and let it run for 30 minutes to determine the peak power consumption while letting the card sit at a stable Windows desktop for 30 minutes to determine the peak idle power consumption. We have also included several other tests as well.

Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the power meter.

To say we are impressed with the power consumption the GT 430 displayed is a huge understatement. We’re not sure if it was the ASUS card’s upgraded components or the possibility that NVIDIA may have been a bit generous with their TDP numbers but seeing it beat the HD 5570 in the power consumption department was shocking.

PLEASE NOTE: the HD 5550 GDDR5 reference model we have has known issues with idle power consumption.

 

[SKYMTL]

Overclocking Results

Overclocking Results

Using the EVGA Precision utility in addition to our usual stability checks, both the core and memory of the GTS 430 1GB were pushed as far as the default voltage would allow. The results for the core were quite good but the lower-spec DDR3 wouldn’t allow its clocks speeds to be increased all that much.


Final Clock Speeds

Graphics Clock: 848Mhz
Processor Clock: 1696Mhz
Memory Clock: 1858Mhz (DDR)

 

[SKYMTL]

Conclusion

Conclusion

The entry level market likely won't that exciting for the vast majority of people reading this article. However, we feel the GT 430 has become a serious contender by injected just the right amount of adrenalin into this somewhat stagnant price point. It really is a pleasant surprise.

NVIDIA didn’t have to reinvent the wheel; they just needed to add some grease to make it turn a bit smoother and that’s exactly what the GT 430 has done. It isn’t a genre defining product by any stretch of the imagination but it delivers at a price point where GeForce cards have been found lacking in the last 18 months. The GT 220 is now replaced with something that is infinitely more capable; one which offers better features, higher performance and yet consumes less power. The addition of lossless audio support as well as a Blu Ray 3D compatible HDMI 1.4 connector is also noteworthy. To us, that is a success from nearly every perspective.

However, there is one nagging detail that sprouted up again and again: the HD 5570. It was released more than eight months ago, still retains its original price of $79 and the release of the GT 430 likely won’t cause AMD to lower their prices one iota. Why? Because the GT 430 maintains the $70 to $80 status quo and really doesn’t put much pressure on the AMD card from a gaming perspective. But framerate pushing is far down the list of what an entry level GPU is meant to accomplish.

When you take off the gaming blinders, a totally different conclusion emerges. Looking at things like support for and performance of GPU computing, it is quite obvious that CUDA is sometimes light years ahead of the competition. ATI’s Stream compute actually works very well but the number of programs that support it is limited and its performance usually leaves much to be desired. NVIDIA has spent the time and loads of money to actively market CUDA to developers and their effort is starting to bear fruit. There are still some concerns with CUDA’s scalability in relation to the GPU’s actual capability but we're hoping that can be taken care of by upcoming driver releases.

When the rubber meets the road, the choice between a HD 5570 a GT 430 will likely come down to either brand preference or a coin toss. They really are well matched up. But as the heart of a media-centric PC, the GT 430 is highly recommendable since it exhibits excellent performance and features for a sub-$99 card and is quite efficient to boot. To us, that makes for a pretty good entry level card.


Pros:

- Acceptable performance for its price
- Surprisingly efficient
- Excellent GPGPU results
- Relatively quiet
- Well priced (according to initial SRP)


Cons:

- Same performance as the HD 5570 but released 8 months later
- Somewhat loud when under heavy load

http://www.hardwarecanucks.com/foru...geforce-gt-430-1gb-review-comment-thread.html​