戴尔 16 Premium CPU 的运行速度比 XPS 16 9640 快 15% 到 20%，但也有一个问题
戴尔 16 Premium现已在全球发货，接替去年的 XPS 16 9640.它用 Arrow Lake-H 和 Blackwell 选项分别取代了 2024 型号上的 Meteor Lake-H 和 Ada Lovelace 选项。与上一代 XPS 16 9640 相比，用户可以期待多线程性能提高 20%。 Core Ultra 7 155H如果配置 酷睿至尊 7 255H.
但问题是，这 20% 的提升仅限于最初的几分钟左右。如下图所示，在戴尔 16 Premium Core Ultra 7 255H 测试机上循环运行 CineBench R15 xT 时，性能分数会随着时间的推移从 3410 分稳步下降到 2900 分左右。相比之下，XPS 16 9640 随着时间的推移会保持稳定。随着节流功能的启动，戴尔 16 Premium 和 XPS 16 9640 的 CPU 性能差距越来越小。
大多数用户甚至可能不会注意到两款笔记本 CPU 性能的微小增减，这使得较旧的 XPS 16 9640 看起来更划算。不过，这也说明了 2024 和 2025 型号之间相对较小的增益和差异。2024 XPS 16 与 2023 XPS 16 相比已经有了很大的进步，因此用户不应该对最新的 2025 版本抱有太高的期望。更多基准测试和比较可参见我们对 2025 型号的评测 这里.
Cinebench R15 多回路运行
Dell 16 Premium DA16250 Intel Core Ultra 7 255H: Ø2953 (2886.63-3410.64)
Dell XPS 16 9640 Intel Core Ultra 7 155H: Ø2707 (2643.06-2817.5)
CPU Performance rating: Percent
Cinebench R23: Multi Core | Single Core
Cinebench R20: CPU (Multi Core) | CPU (Single Core)
Cinebench R15: CPU Multi 64Bit | CPU Single 64Bit
Blender: v2.79 BMW27 CPU
7-Zip 18.03: 7z b 4 | 7z b 4 -mmt1
Geekbench 6.5: Multi-Core | Single-Core
Geekbench 5.5: Multi-Core | Single-Core
HWBOT x265 Benchmark v2.2: 4k Preset
LibreOffice : 20 Documents To PDF
R Benchmark 2.5: Overall mean
Cinebench R23: Multi Core | Single Core
Cinebench R20: CPU (Multi Core) | CPU (Single Core)
Cinebench R15: CPU Multi 64Bit | CPU Single 64Bit
Blender: v2.79 BMW27 CPU
7-Zip 18.03: 7z b 4 | 7z b 4 -mmt1
Geekbench 6.5: Multi-Core | Single-Core
Geekbench 5.5: Multi-Core | Single-Core
HWBOT x265 Benchmark v2.2: 4k Preset
LibreOffice : 20 Documents To PDF
R Benchmark 2.5: Overall mean
| CPU Performance rating | |
| Dell 16 Premium DA16250 | |
| Average Intel Core Ultra 7 255H | |
| Dell XPS 16 9640 | |
| Cinebench R23 / Multi Core | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Average Intel Core Ultra 7 255H (16105 - 22578, n=15) | |
| Cinebench R23 / Single Core | |
| Average Intel Core Ultra 7 255H (1874 - 2152, n=15) | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Cinebench R20 / CPU (Multi Core) | |
| Dell 16 Premium DA16250 | |
| Average Intel Core Ultra 7 255H (6356 - 8860, n=14) | |
| Dell XPS 16 9640 | |
| Cinebench R20 / CPU (Single Core) | |
| Average Intel Core Ultra 7 255H (748 - 826, n=14) | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Cinebench R15 / CPU Multi 64Bit | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Average Intel Core Ultra 7 255H (2414 - 3444, n=14) | |
| Cinebench R15 / CPU Single 64Bit | |
| Average Intel Core Ultra 7 255H (258 - 322, n=15) | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Blender / v2.79 BMW27 CPU | |
| Average Intel Core Ultra 7 255H (157 - 293, n=14) | |
| Dell XPS 16 9640 | |
| Dell 16 Premium DA16250 | |
| 7-Zip 18.03 / 7z b 4 | |
| Dell XPS 16 9640 | |
| Dell 16 Premium DA16250 | |
| Average Intel Core Ultra 7 255H (42155 - 68652, n=15) | |
| 7-Zip 18.03 / 7z b 4 -mmt1 | |
| Average Intel Core Ultra 7 255H (4589 - 6385, n=15) | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Geekbench 6.5 / Multi-Core | |
| Dell 16 Premium DA16250 | |
| Average Intel Core Ultra 7 255H (11979 - 17489, n=15) | |
| Dell XPS 16 9640 | |
| Geekbench 6.5 / Single-Core | |
| Average Intel Core Ultra 7 255H (2712 - 2925, n=15) | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Geekbench 5.5 / Multi-Core | |
| Dell 16 Premium DA16250 | |
| Average Intel Core Ultra 7 255H (11348 - 17218, n=14) | |
| Dell XPS 16 9640 | |
| Geekbench 5.5 / Single-Core | |
| Average Intel Core Ultra 7 255H (1938 - 2121, n=14) | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| HWBOT x265 Benchmark v2.2 / 4k Preset | |
| Dell 16 Premium DA16250 | |
| Dell XPS 16 9640 | |
| Average Intel Core Ultra 7 255H (13.7 - 25.8, n=14) | |
| LibreOffice / 20 Documents To PDF | |
| Average Intel Core Ultra 7 255H (41.2 - 61.6, n=14) | |
| Dell XPS 16 9640 | |
| Dell 16 Premium DA16250 | |
| R Benchmark 2.5 / Overall mean | |
| Dell XPS 16 9640 | |
| Dell 16 Premium DA16250 | |
| Average Intel Core Ultra 7 255H (0.4233 - 0.515, n=14) | |
* ... smaller is better
