Ask shopping questions and we'll answer in our next live Webcast. Visit our Gadget Gift Guide.
Find a Review



Processor Face-off: Intel Santa Rosa vs. Penryn

Share Print
by Jeffrey L. Wilson on February 11, 2008

penryn_vs_santa_rosa_hThe first wave of Penryn notebooks have begun rolling out to consumers (such as the HP Pavillion HDX and the Toshiba Satellite X205-SLi4), and with them comes the promise of machines that are more efficient in terms of both processing power and power consumption.
At the heart of the new Penryn processors are 45-nanometer transistors whose miniscule size enables Intel to pack more of them into each Core 2 Duo processor for extra processing punch. Plus, they're built of new “Hi-K” materials that significantly reduce electrical leakage, resulting in 30-percent lower power consumption than typical silicon-based transistors. It all sounds groovy on paper, but as tech-heads, we wondered how these specs translate into real world performance. Thanks to Intel, we were able to find out.

Intel provided us with a pair of similarly-configured Dell Latitude D630 systems (2GB RAM, three 7,200-rpm 160GB hard drives, 9-cell batteries, Windows Vista Ultimate operating systens) that only differed in their individual processors. One featured a 2.6-GHz Intel Core 2 Duo Santa Rosa-powered engine; the other featured a 2.6-GHz Intel Core 2 Duo Penryn chip. Also included were several benchmarks that we used to test Penryn's mettle. Here's how the new chipset fared when matched up against Santa Rosa.
 

Test 1: Sony Vegas Movie Studio Platinum

This benchmark simulates someone transferring a 24 minute and 46 second high-definition video clip (1,440 x 1,080 resolution, 29.9 frames per second, 4.7GB) to a notebook running on battery power, and then encoding it into a 1.4GB DivX file.  With the Santa Rosa machine, the process took 65.2 minutes; with the Penryn system, the process took 43.2 minutes – a whopping 20-minute difference.
 

Test 2: TMPGEnc Xpress

This benchmark simulates converting a high-resolution (1,440 x 1,080), 1 minute and 16 second video clip with VBR encoding to an MPEG-4 file with 25,000Kb/s CBR encoding. With the Santa Rosa machine, the process took 4.7 minutes as compared to 3.2 with the Penryn model.


Test 3: Virtual Dub

This benchmark simulates converting a 66.2MB, 720 x 480-pixel resolution home video into DivX format. With the Santa Rosa notebook, the process took 43.9 seconds as compared to 28.4 seconds with the Penryn model.
 

Test 4: Premiere Pro CS3

This benchmark measures the time it takes to render a non-realtime preview of the first 7 seconds of 1080p video footage. With the Santa Rosa machine, the process took 49.4 seconds as compared to 39.1 with Penryn.
 

Test 5:  Word and PowerPoint

This benchmark simulates converting a 50 page, 6.6MB presentation file into a 9MB XPS file that can be viewed from any web browser. It also simulates someone comparing large 1,000+ page documents, and creating a merged document containing only the differences between the original two documents. With Santa Rosa system, the XPS conversion took 1 minute and 9 seconds, while Word Merge took 46.1 seconds. On the Penryn system, the XPS conversion took 1 minute and 2 seconds, while Word Merge took 47.2 seconds.


Test 6:  MS Excel Monte Carlo

The Black-Scholes model is used to calculate a theoretical call and put price using the five key determinants of an option's price: stock price, strike price, volatility, time to expiration, and short-term (risk free) interest rate. The workload is execution of approximately 300,000 iterations  of Monte Carlo simulation using the Black-Scholes basic option pricing formula  in Microsoft Excel 2007.  In addition, the workload uses Excel lookup functions to compare the put price from the model with the historical market price for 50,000 rows to understand the convergence.  The input file is a 70.1 MB spreadsheet. The Santa Rosa notebook took 43.3 seconds to complete the scenario, while the Penryn notebook took 39 seconds.
 

Test 7: Adobe Photoshop Lightroom 1.3

This benchmark simulates converting 200 photos (830MB in size, total) to a 480 x 360 resolution and 160 pixels per inch, for viewing on portable devices like the iPhone. The photos are then posted to a web gallery online. The Santa Rosa machine took 5 minutes to resize the photos, and 3.8 minutes to post them online. The Penryn system took 4.3 minutes to resize the photos, and 4.1 minutes to post them online.
 

Test 8: Cinebench 9.5 and Cinebench R10

This benchmark simulates 3D content creation. The Santa Rosa notebook took 27 seconds, and 2 minutes and 55 minutes to render a scene in Cinebench 9.5 and Cinebench R10, respectively; the Penryn notebook took 25 seconds, and 2 minutes and 43 minutes to render the same scenes.
 

Test 9: Battery Life with MobileMark 2007

When we ran MobileMark 2007, the Penryn-powered Dell Latitude D630 lasted 5 hours and 32 minutes with its Wi-Fi signal on, which increased by over half an hour to 6 hours and 03 minutes with the wireless connection deactivated. The Santa Rosa system saw a 5 hour and 19 minute runtime with the Wi-Fi on (13 minutes shorter than its Penryn counterpart) and 5 hours and 49 minutes with it off (14 minutes shorter). 
Although the difference in endurance isn't staggering, road warriors will appreciate every extra amount of juice they can get out of their systems. We also ran 3DMark03, which saw scores of 1,776 and 1835 from the Santa Rosa and Penryn processors, respectively.

Final Verdict: Penryn vs. Santa Rosa


Except for the Adobe Premier Pro CS3, Sony Vegas Movie Studio Platinum, and Virtual Dub (which saw the Penryn processor slice computing times by 10.3 seconds, 22 minutes, and 15.5 seconds, respectively), the Santa Rosa and Penryn results were quite comparable; it seems that Penryn's potential shines through the most for those that do heavy video work. The average consumer may not immediately see the benefit of Penryn, but the increased battery life that it provides should appeal to all users.

Market Place

Featured Sponsors