 | |||
|
|
|
|
|
 |
NVIDIA GeForce 9500 GT Performance Preview July 29, 2008 Brandon Sandman Bell |
|
 |
|
| |
Summary: Based on NVIDIA's G96 GPU, the GeForce 9500 GT is taking the mainstream graphics market by storm starting today. Join us as we examine its performance at stock and OC'ed speeds against six GPUs ranging from the 8500 GT and Radeon 3650, up to the latest GeForce 9600 GT. Does NVIDIA have a winner on their hands? Only one way to find out -- read the article!
 GeForce 9500 GT Performance Preview | Page:: ( 1 / 13 ) |
[image]
| <% print_image("01"); %> |
After getting off to a sluggish start, the DirectX 10 gaming era is now in full swing. Since the end of last summer, we’ve witnessed the arrival of dozens of new DirectX 10 games, and many more are in the works for debut later this year. Even blockbuster console games like Gears of War and Assassin’s Creed have been ported to the PC with enhanced DirectX 10 graphics; expect this trend to continue as well.
But DirectX 10 isn’t the only reason why we’re excited about the current generation of GPUs. Today’s latest GPUs can be used for much more than just playing games. Video features such as Blu-ray and DVD playback are old hand by now, while the addition of advanced post-processing effects such as dynamic contrast enhancement have given the GPU features that many high-end dedicated Blu-ray players don’t support.
[image]
| <% print_image("02"); %> | <% print_image("03"); %> |
We’re approaching a new era in 3D graphics where even more tasks that were traditionally handled by the CPU can now be tackled by the GPU, with improved performance. Whereas today transcoding one 1080p video can be a time-consuming process for the CPU, in the near future you’ll be able to do it in half the time on the GPU, or transcode multiple 1080p videos simultaneously. Adobe even plans to bring GPU acceleration to Photoshop later this year. (And of course NVIDIA doesn’t want you to forget physics.)
Eventually other common tasks such as audio encoding and 3D rendering could potentially be handled by the GPU.
For the average joe consumer, what’s really exciting is you don’t have to fork over hundreds of dollars to get this capability. Any DX10 card will do. Thanks to the latest round of GPU price cuts, this means you can spend anywhere from $50-$500 to get started. The most exciting developments are arguably occurring in the sub-$200 space. Here both ATI and NVIDIA offer a wide range of products: NVIDIA alone has offered ten different GPUs in this price range over the course of the last year.
And with today’s introduction of the GeForce 9500 GT, you can now add an 11th.
[image]
| <% print_image("04"); %> | <% print_image("05"); %> |
Slated just below the GeForce 9600 GT, NVIDIA’s GeForce 9500 GT replaces the GeForce 8600 GPUs in NVIDIA’s lineup. Its G96 GPU sports 32 stream processors and is built on TSMC’s 65-nm and 55-nm manufacturing process, with the chip containing 314 million transistors total. In comparison, the G84 chip powering NVIDIA’s GeForce 8600 series was built on TSMC’s larger 80-nm manufacturing process and contained 289 million transistors. GeForce 9500 GT cards are expected to retail for about the same price as today’s GeForce 8600 GT cards. We haven’t been given an exact MSRP, but expect a price tag in the low $80s range.
But how does NVIDIA’s new G96 GPU stack up in comparison to its predecessors? Let’s take a closer look under the hood?
| G96 Graphics Core | Page:: ( 2 / 13 ) |
[image]
| <% print_image("06"); %> | <% print_image("07"); %> |
On paper, NVIDIA’s G96 GPU utilized in the GeForce 9500 GT looks awfully similar to the G84 GPU used in the GeForce 8600 series. Both GPUs feature the same number of stream processors (32), ROPs (8) and texture address and filtering units (16). Both chips also utilize a 128-bit memory interface to either GDDR3 or DDR2 memory. At first glance the only notable difference is the G96 GPU’s smaller manufacturing process.
However NVIDIA has incorporated several tweaks into their G9x GPUs in comparison to the G8x generation. For starters there’s PCI Express 2.0. PCIe 2.0 offers double the bandwidth of PCIe 1.1; 8.0GB/sec in each direction, providing a total of 16GB/sec of total bandwidth
In addition, G96 (like G92 before it), boasts improved color and z-compression over G84. This allows the GPU to make more efficient use of its available memory bandwidth, and should help the most at high resolutions, particularly once AA is applied.
Where did the rest of the extra transistors go? Besides the aforementioned improvements, G96 also adds support for DisplayPort displays. We’ve also been told that transistor counts can vary between each manufacturing process, so expect some variance there. All told G96 should offer clock-for-clock performance improvements over G84, and thanks to its smaller process the chip is cheaper for NVIDIA to produce, with lower power consumption as well. Here are the complete specs on G96, as well as its predecessors, G84 (GeForce 8600 series) and G86 (GeForce 8500 GT):
| Mainstream GeForce Comparison | |||||
| GeForce 9500 GT 256MB/512MB GDDR3 | GeForce 9500 GT 512MB DDR2 | GeForce 8600 GTS | GeForce 8600 GT | GeForce 8500 GT | |
| # of Stream Processors | 32 | 32 | 32 | 32 | 16 |
| Core Clock Speed | 550MHz | 550MHz | 675MHz | 540MHz | 450MHz |
| Shader Clock Speed | 1.4GHz | 1.4GHz | 1.45GHz | 1.19GHz | 900MHz |
| Texture Address/Filtering Units | 16/16 | 16/16 | 16/16 | 16/16 | 8/8 |
| Texture Fill-Rate | 8.8 Gigatexels/sec | 8.8 Gigatexels/sec | 10.8 Gigatexels/sec | 8.6 Gigatexels/sec | 3.6 Gigatexels/sec |
| Memory Clock Speed | 800MHz (1.6GHz effective) | 500MHz (1.0GHz effective) | 1.0GHz (2.0GHz effective) | 700MHz (1.4GHz effective) | 400MHz (800MHz effective) |
| Memory Interface | 128-bit | 128-bit | 128-bit | 128-bit | 128-bit |
| Memory Bandwidth | 25.6GB/sec | 16GB/sec | 32GB/sec | 22.4GB/sec | 12.8GB/sec |
| PCI Express Interface | 2.0 | 2.0 | 1.1 | 1.1 | 1.1 |
| Max Board Power Draw | 50W | 50W | 71W | 43W | N/A |
| HDCP | Yes | Yes | Yes | Yes | Optional |
| |||||
Notes
As you can see, NVIDIA is providing two different GeForce 9500 GT SKUs: one with GDDR3 memory, and one SKU with DDR2 RAM. The graphics core and shader clocks are the same on both SKUs, with the GDDR3 board shipping with faster memory.
NVIDIA has also unofficially developed a third reference design SKU with OC’ed clocks. This is the first time we can recall NVIDIA doing this. The so-called GeForce 9500 GT OC runs at 600MHz core while the stream processors operate at 1500MHz. The board’s memory is then clocked at 1.0GHz (2.0GHz effective), matching the clock speed of the GeForce 8600 GTS. This is the board NVIDIA provided for our testing today, although we also underclocked the board to the factory settings for our performance benchmarks as well.
Normally we’d be hesitant to test an OC’ed reference design (retail cards are obviously different), but considering the enormous overclocking results we obtained with our G96 sample, we’re actually a little surprised the OC’ed clock speeds aren’t the stock speeds. We managed to hit OC’ed speeds of 750MHz core, 1200MHz memory, and 1875MHz on the stream processors! Based on this, we’re pretty confident that all the usual suspects should have factory OC’ed GeForce 9500 GT cards, and some of them may actually go well beyond NVIDIA’s own GeForce 9500 GT OC.
Board analysis
NVIDIA’s GeForce 9500 GT reference board design is quite a departure from previous GeForce cards we’ve tested lately like the 9800 GTX+ and GTX 260/280. The card measures in with a diminutive 7” PCB and single-slot cooling. Since the board draws only 50W peak power consumption, no external power connector is necessary, instead the board draws its power solely from the PCIe interface.
For cooling the GPU, NVIDIA borrows the same heatsink/fan design that was first integrated on the GeForce 8600 GT. The cooling unit consists of a simple aluminum heatsink which is paired with a near-silent cooling fan.
[image]
| <% print_image("08"); %> | <% print_image("09"); %> |
As always with lower-end cards, NVIDIA’s board partners are free to make modifications for their own final 9500 GT cards, so we expect all kinds of derivatives to pop up over the course of the next few weeks. Already Palit has announced that they will be producing a passively cooled GeForce 9500 GT that runs at the stock GeForce 9500 GT clocks. Based on their history, ASUS, Gigabyte, and MSI will likely have passively cooled GeForce 9500 GT cards of their own shortly as well. These passively cooled 9500 GT cards would obviously be geared towards the HTPC crowd.
So now that you know more about the GeForce 9500 GT and its new G96 graphics core, how does it stack up against previous GeForce offerings as well as the competition in terms of performance? Let’s find out shall we?
| System Setup | Page:: ( 3 / 13 ) |
Intel Core 2 Duo E6750
ASUS P5E3 Deluxe
4GB Crucial Ballistix DDR3
Albatron GeForce 8600 GTS
Foxconn GeForce 8600 GT
EVGA e-GeForce 8500 GT
Zogis GeForce 9600 GT
ForceWare 175.19
GeForce 9500 GT
ForceWare 177.72
ATI Radeon 3650 512MB
ATI Radeon 3850 512MB
Catalyst 8.8
300GB Western Digital Caviar SE
Windows Vista Ultimate 64-bit w/Service Pack 1
Benchmarks
Company of Heroes 1.71
F.E.A.R. 1.08
Elder Scrolls IV: Oblivion 1.2
Call of Duty 4 1.4
Half-Life 2 Episode Two
Lost Planet
Crysis 1.2
Game Settings Used
[image]
| <% print_image("10"); %> | <% print_image("11"); %> | <% print_image("12"); %> |
| <% print_image("13"); %> | <% print_image("14"); %> | <% print_image("15"); %> |
| <% print_image("16"); %> | <% print_image("17"); %> | <% print_image("18"); %> |
| <% print_image("19"); %> | <% print_image("20"); %> |
| F.E.A.R. Performance | Page:: ( 4 / 13 ) |
| Company Of Heroes DX10 | Page:: ( 5 / 13 ) |
| Quake Wars | Page:: ( 6 / 13 ) |
| HL2 Episode 2 | Page:: ( 7 / 13 ) |
| Lost Planet DX10 | Page:: ( 8 / 13 ) |
| Call Of Duty 4 | Page:: ( 9 / 13 ) |
| Crysis DX10 | Page:: ( 10 / 13 ) |
Crysis High – Direct3D
| BioShock DX10 | Page:: ( 11 / 13 ) |
| Overclocking | Page:: ( 12 / 13 ) |
| Conclusion | Page:: ( 13 / 13 ) |
Thanks to its smaller manufacturing process, GeForce 9500 GT is considerably cheaper for NVIDIA to produce, while at the same time the GPU performs closest to the GeForce 8600 GTS. The GeForce 8600 GTS’ G84 GPU itself is not only costlier, its board design is as well, requiring more elaborate cooling and additional board-level components for power delivery. In comparison the GeForce 9500 GT is much simpler, unlike the 8600 GTS the board doesn’t require an external power connector, and don’t forget the enhancements NVIDIA has added to their G9x GPUs to improve clock-for-clock performance, as well as the addition of newer features like DisplayPort and PCIe 2.0.
Fundamentally G96 is an evolutionary progression of G84, just as G92 was for G80. NVIDIA isn’t drastically breaking any new ground when it comes to the architecture or its performance, on average the GeForce 9500 GT OC ran around 6-9% faster than the GeForce 8600 GTS, with the notable exceptions being Company of Heroes DX10 and Crysis DX10. In these games the difference was much greater, with the 9500 GT OC running 12-20% faster than the GeForce 8600 GTS. Meanwhile the bone stock GeForce 9500 GT generally ran about 5-8% slower than the GeForce 8600 GTS, although the stock 9500 GT managed to outrun the 8600 GTS in both Company of Heroes and Crysis.
Based on all this, we consider the GeForce 9500 GT OC to be the spiritual successor of the 8600 GTS, while the stock 9500 GT is replacing the 8600 GT. According to NVIDIA, the GeForce 8500 GT and GeForce 8400 GS will continue to be produced while the 8600s will slowly fade away.
And what about the competition? As of today the GeForce 9500 GT is officially going head-to-head with ATI’s Radeon HD 3650, but we all know that ATI is currently putting the finishing touches on their upcoming next-generation RV730 GPU. This GPU will have its sights set directly at the GeForce 9500 GT, as RV730 will presumably go into Radeon 4650 cards when it launches later this summer. If ATI is able to deliver a coup similar to the one they achieved with RV770, RV730 could put NVIDIA’s 9500 GT back on its heels.
Of course, at the same time based on our OC’ing results with G96, it’s entirely possible that NVIDIA is holding back right now, as it certainly appears they may have tons of frequency headroom that’s sitting untapped right now in G96. If RV730 is indeed a strong competitor, they could then respond with a 650MHz or perhaps even a 700MHz GeForce 9500 GT+. Only NVIDIA knows for sure.
What we do know is that NVIDIA is busily transitioning their entire GeForce lineup to 55-nm. With the 9500 GT launch now behind them, NVIDIA is now focusing on 55-nm variants of the GeForce 9800 GT, followed by the GeForce 9600 GT, and ultimately the GeForce GTX 260 and GTX 280. If all goes well for NVIDIA, their entire lineup should be at 55-nm by the end of September, if not slightly sooner.
With so many GPUs in the works from both ATI and NVIDIA, we’re going to be awfully busy the next few months…