Toshiba Satellite X205-S9359 Take Two: Displays and Drivers
- Posted in
- Laptops
Driver Performance Comparison
Having managed to install the 163.44 drivers, we were also curious to see how they compared in terms of performance to the official drivers. We tried using hacked drivers on the Alienware m9750 with pretty disappointing results, but the use of SLI on that notebook may have been a significant contributing factor. Our test system is unchanged from the previous article, other than using the newer graphics drivers.
We'll provide benchmark charts with additional details about the performance in various applications and games in a moment, but for those looking for a quick overview of how the 163.44 drivers affect performance on the X205, here's the average performance change for a number of games; we tested that 1280x800, 1440x900, and 1680x1050.
Many of the titles show insignificant performance changes - anything less than 3%. The overall performance change averages out to 3%, which may at first seem like the hacked drivers aren't worth the trouble. However, we didn't include the Company of Heroes DirectX 10 result in that figure, as it seriously skewed the overall performance change. If that particular game is any indication of the status of DirectX 10 drivers and games, there's clearly a lot of potential for future improvements.
Looking at the remaining titles, most show moderate to slightly noticeable performance increases. Antialiasing seems to benefit more from the updated driver in several applications; however, it's also important to mention that in several titles enabling 4xAA simply isn't practical without dropping to a much lower resolution. F.E.A.R. for example only averages 26/27 FPS with 4xAA at 1680x1050, versus 48/51 without it. Some people might actually feel that antialiasing is more important than a high resolution, but we would strongly disagree - especially when using LCDs where running at the native resolution is optimal. Battlefield 2 and the Half-Life 2 games on the other hand are easily playable with 4xAA, making the results pertinent. Half-Life 2: Lost Coast and Half-Life 2: Episode One are interesting in that both show a significant performance drop without antialiasing, but they also show an even larger performance gain with antialiasing enabled. We would expect that trend to continue with other HL2 engine games like Counter Strike: Source and the original HL2.
The bottom line, of course, is that at least one game - Bioshock - simply refused to run without using the hacked/updated drivers. We can be pretty certain that it won't be the last game to require something more recent than the official 101.68 drivers. Will NVIDIA and/or Toshiba manage to release updated drivers before other titles encounter difficulties? More likely than not the answer is yes and no: yes, they will get new drivers out, but at some point even those new drivers are going to have issues and require another driver update. For anyone using a laptop that isn't part of a corporate environment, the ability to install updated graphics drivers rather than waiting a few extra months for WHQL certification could certainly be helpful.
Having managed to install the 163.44 drivers, we were also curious to see how they compared in terms of performance to the official drivers. We tried using hacked drivers on the Alienware m9750 with pretty disappointing results, but the use of SLI on that notebook may have been a significant contributing factor. Our test system is unchanged from the previous article, other than using the newer graphics drivers.
| Toshiba X205-S9359 Tested Configuration | |
| Processor | Core 2 Duo T7300 (2.00GHz 4MB 800FSB) |
| Chipset | Intel GM965 + ICH8-ME |
| Memory | 2x1024MB DDR2 SO-DIMMs (Hyundai HYMP512S64CP8-T5)Tested at DDR2-667 5-5-5-15 |
| Graphics | NVIDIA GeForce 8700M GT 512MB w/255MB TurboCache |
| Display | 17" WSXGA+ TruBright (1680x1050)LG Philips LP171WE3 (Jan 2007) |
| Hard Drive | 2 x 160GB 5400RPM 8MB SATA(Toshiba MK1637GSX) |
| Optical Drive | Toshiba HD DVD TS-L802A |
| Audio | Realtek ALC268 HD Audio |
| Battery | 9-Cell 65WHr |
| Operating System | Windows Vista Ultimate 32-bit |
We'll provide benchmark charts with additional details about the performance in various applications and games in a moment, but for those looking for a quick overview of how the 163.44 drivers affect performance on the X205, here's the average performance change for a number of games; we tested that 1280x800, 1440x900, and 1680x1050.
| Toshiba Satellite X205-S9359 (8700M GT 512MB) Driver Performance Comparison |
|||
| Game | 1680x1050 101.68 | 1680x1050 163.44 | Average Change |
| Battlefield 2 0xAA | 69.71 | 71.28 | 1.61% |
| Battlefield 2 4xAA | 36.53 | 39.39 | 7.20% |
| Far Cry 0xAA | 49.19 | 48.87 | -0.20% |
| Far Cry 4xAA | 34.49 | 34.43 | 0.19% |
| FEAR | 48.00 | 51.00 | 8.25% |
| FEAR 4xAA | 26.00 | 27.00 | 3.96% |
| HL2: Lost Coast 0xAA | 47.41 | 42.39 | -14.39% |
| HL2: Lost Coast 4xAA | 28.49 | 31.67 | 25.46% |
| HL2: Episode One 0xAA | 71.20 | 65.25 | -7.68% |
| HL2: Episode One 4xAA | 43.55 | 49.85 | 14.51% |
| Quake 4 0xAA | 66.40 | 65.40 | -1.10% |
| Quake 4 4xAA | 37.80 | 37.10 | 1.35% |
| Oblivion | 18.19 | 18.01 | -4.04% |
| Supreme Commander | 14.04 | 15.01 | 2.94% |
| STALKER | 15.40 | 16.94 | 8.74% |
| Company of Heroes DX9 | 29.50 | 30.60 | 3.09% |
| Average Performance Change | 3.12% | ||
| DX10 Comparison | |||
| Game | 1680x1050 101.68 | 1680x1050 163.44 | Average Change |
| Company of Heroes DX10 | 4.20 | 11.90 | 117.91% |
Many of the titles show insignificant performance changes - anything less than 3%. The overall performance change averages out to 3%, which may at first seem like the hacked drivers aren't worth the trouble. However, we didn't include the Company of Heroes DirectX 10 result in that figure, as it seriously skewed the overall performance change. If that particular game is any indication of the status of DirectX 10 drivers and games, there's clearly a lot of potential for future improvements.
Looking at the remaining titles, most show moderate to slightly noticeable performance increases. Antialiasing seems to benefit more from the updated driver in several applications; however, it's also important to mention that in several titles enabling 4xAA simply isn't practical without dropping to a much lower resolution. F.E.A.R. for example only averages 26/27 FPS with 4xAA at 1680x1050, versus 48/51 without it. Some people might actually feel that antialiasing is more important than a high resolution, but we would strongly disagree - especially when using LCDs where running at the native resolution is optimal. Battlefield 2 and the Half-Life 2 games on the other hand are easily playable with 4xAA, making the results pertinent. Half-Life 2: Lost Coast and Half-Life 2: Episode One are interesting in that both show a significant performance drop without antialiasing, but they also show an even larger performance gain with antialiasing enabled. We would expect that trend to continue with other HL2 engine games like Counter Strike: Source and the original HL2.
The bottom line, of course, is that at least one game - Bioshock - simply refused to run without using the hacked/updated drivers. We can be pretty certain that it won't be the last game to require something more recent than the official 101.68 drivers. Will NVIDIA and/or Toshiba manage to release updated drivers before other titles encounter difficulties? More likely than not the answer is yes and no: yes, they will get new drivers out, but at some point even those new drivers are going to have issues and require another driver update. For anyone using a laptop that isn't part of a corporate environment, the ability to install updated graphics drivers rather than waiting a few extra months for WHQL certification could certainly be helpful.
Originally published at https://www.anandtech.com/show/2317 (Wayback Machine).