Independent tests find AMD's Opterons more power efficient than Intel's Xeons in 36 out of 57 cases

Independent tests conducted by Neal Nelson & Associates, an independent computer performance consulting firm, have found that AMD's Opteron-based servers deliver better power efficiency than Intel Xeon-based counterparts.

A quote from the laboratory test report reads:

The tests were performed on servers configured with 2, 4, 6, and 8 gigabytes of main memory at various transaction processing load levels. The results show that for certain configurations and at certain load levels the Intel Xeon based server was 2.4 to 11.7 percent more power efficient while in other cases the AMD Opteron based server was 9.2 to 23.1 percent more power efficient. In addition, when the systems were idle and waiting for transactions to process, the AMD server was 30.4 to 53.1 percent more power efficient.

AMD's Opteron-based servers were found to provide more power efficiency with increasing memory size. Also, the tests concluded that while Intel's Xeon was more efficient for calculation intensive workloads 5.0 to 5.5 percent, AMD Opterons were more power efficient at I/O intensive operations by 18.4 to 18.6 percent.

An article at EDN quotes that Neal Nelson, the firm's founder, undertook the tests to independently verify Intel CEO Paul Otellini's claim concerning Intel's lead in power efficiency. An Intel representative disputed the report, saying that the test does not consider the performance of Intel's latest Xeons, or quad cores.

More links:

Research firm claims AMD beats Intel on power efficiency (The Inquirer)

AMD Beats Intel in 36 of 57 Power Efficiency Tests by Neal Nelson (dbusinessNews)

georgeou Oh Christ, this is a fraud. Not only did he rig the test, his test is fundamentally worthless. Go look at his latest results in his "white paper" where a single 2GHz system with 1GB RAM gets IDENTICAL transactions per minute performance compared to a dual 2.33 GHz system with 16GB RAM. Wow, and he has the gall to slam SPECpower.


Well, there is no surprise. Intel was never good in IO operations, internal registry content manipulation and cache structuring...


This is pretty standard Intel misrepresents there processor,not a sole in the media say a word,those of us that know have no power to inform,and by the time the media catches on,in this case a year later,Intel has produced a product that could make there claim true.They have been doing it for years.They attempt to make what they said for a year true with a product they released last week