PowerPC Setup Benchmarks | |
Sven Negrassus - 27. März 1999 | |
Diese und die folgenden Seiten vergleichen unterschiedliche Konfigurationen einer G3-Upgradekarte bezüglich Prozessor MHz, Cachegrösse und -speed sowie Bustakt. Die Tests wurden mit einer Met@box joeCARD 400 durchgeführt, die für die meisten Tests auf 300 MHz heruntergetaktet wurde. Basisrechner ist ein Power Macintosh 8500 mit 96MB RAM (interleaved) und installiertem MacOS 8.5.1. | This and the following pages compare different settings of a G3 upgrade card regarding processor MHz, cache size & speed and bus speed. The tests were run on a Met@box joeCARD 400, downclocked to 300 MHz for most tests. Test machine is a Power Macintosh 8500 with 96MB interleaved RAM and MacOS 8.5.1 installed. |
Prozessor MHz | Bus Speed | Cache Size | Cache Speed |
Prozessor MHz | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Technische Daten | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Konfiguration: | Prozessor MHz | L2-Cache | L2-MHz | Bus MHz | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
joeCARD 300/150/50/1MB | 300.0 100% | 1024k | 150.0 2:1 | 50.0 100%
joeCARD 350/175/50/1MB
| 350.1 117%
| 1024k
| 175.0 2:1
| 50.0 100%
| joeCARD 400/200/50/1MB
| 400.0 133%
| 1024k
| 200.0 2:1
| 50.0 100%
| Benchmarks
|
| Macbench 5.0
| CPU
| FPU
| joeCARD 300 MHz
| 987 100.0%
| 1002 100.0%
| joeCARD 350 MHz
| 1143 115.8%
| 1164 116.2%
| joeCARD 400 MHz
| 1316 133.3%
| 1333 133.0%
| Real World Tests
| NeuroNet
| POVray
| joeCARD 300 MHz
| 356 sec. 100.0%
| 245 sec. 100.0%
| joeCARD 350 MHz
| 306 sec. 116.4%
| 209 sec. 117.2%
| joeCARD 400 MHz
| 268 sec. 132.8%
| 184 sec. 133.2%
| Real World Tests
| Cinema 4DXL
| Cinebench SP Index
| joeCARD 300 MHz
| 691 sec. 100.0%
| 4.07 100.0%
| joeCARD 350 MHz
| 585 sec. 118.1%
| 4.71 115.7%
| joeCARD 400 MHz
| 512 sec. 135.0%
| 5.39 132.4%
| Real World Tests
| PhotoShop Actions (38MB Image)
| Norton Gfx Bench @1024x768x16bit
| joeCARD 300 MHz
| 223 sec. 100.0%
| 217 100.0%
| joeCARD 350 MHz
| 209 sec. 106.7%
| 224 103.2%
| joeCARD 400 MHz
| 196 sec. 113.8%
| 228 105.1%
| Real World Tests
| VirtualPC 2.1 / Winbench 97 32 Bit (W95)
| VirtualPC 16 Bit
| joeCARD 300 MHz
| 126 100.0%
| 85.9 100.0%
| joeCARD 350 MHz
| 133 105.6%
| 93.6 109.0%
| joeCARD 400 MHz
| 144 114.3%
| 100.0 116.4%
|
| Memory Throughput
|
|
| Fazit: Während viele Applikationen entsprechend der höheren Taktrate um entsprechend viele Prozente schneller arbeiten (in den Beispielen 17 bzw. 33%), gibt es einige Anwendungen, die weniger von den mehr MegaHertz profitieren. Beispiele dazu sind PhotoShop (mit grossen Bildern) oder VirtualPC (Pentium MMX-Emulation) sowie sehr grafiklastige Applikationen.
| Round up: Many applications work faster that many percent as the processor clock is set higher (17 resp. 33% in our examples), but there are some kinds of programs that do not benefit that much from a faster CPU. Examples are PhotoShop (processing on large images) or VirtualPC (emulating a Pentium MMX) as well as graphics intense applications.
|
| Next: Bus Speed * bus speed
| |
Letzte Änderung * Last Update: %LastFileUpdateDate% |