Xeon and Opteron comparisons: correction

WHEN THIS story was originally published there were percentage errors in the performance lead that Opteron had over Xeon. Even though the test result data was correct, the calculation used to determine Opteron's lead was wrong. So Opteron's overall performance lead had been understated by a very significant degree.

Specifically, for SPECint_rate2000, SPECfp_rate2000, SPECweb99_ssl, SPECweb99 (Red Hat CA2), and SPECjbb2000, I had stated Opteron's lead over Xeon as being 19%, 55%, 48%, 38%, and 17% respectively. Four of the five results were wrong. The percentages should have read as 19%, 107%, 75%, 57%, and 20% respectively.

We apologise for these errors and include the corrected relevant portions to reflect how the Opteron platform truly performs.

The car dealership scenario has also been updated to reflect these results as well.

I have also increased the accuracy of the percentage numbers that follow the test data results, so that those who compare the raw data with the percentage numbers will come out with the same results.

We thank our alert readers for bringing these errors to our attention.

SPECint_rate2000 4P Servers
Opteron 846 - 56.7 (116.91%)
Opteron 844 - 48.5 (100.00%)
Xeon MP 2.8 GHz (2 MB L3) - 47.6 (98.14%)
Xeon MP 2.0 GHz (2 MB L3) - 34.7 (71.55%)

Opteron is showing a 19% performance lead in the first benchmark discipline, and it's doing so without using PC3200 memory.

SPECfp_rate2000 4P Servers
Opteron 846 - 52.5 (106.71%)
Opteron 844 - 49.2 (100.00%)
Xeon MP 2.8 GHz (2 MB L3) - 25.4 (51.63%)
Xeon MP 2.0 GHz (2 MB L3) - 20.2 (41.06%)

Xeon is again found wanting, as Opteron punches out a mammoth 107% lead. Opteron is still using slower PC2700 memory.

SPECweb99_ssl 4P Servers
Opteron 848 - 3800 (111.80%)
Opteron 846 - 3399 (100.00%)
Xeon MP 2.8 GHz (2 MB L3) - 2177 (64.05%)

Opteron is hitting a 75% lead in this category, which happens to be another double digit embarrassment for Xeon. In this configuration, Opteron could make do with PC2100 memory.

SPECweb99 (Red Hat CA2) 4P Servers
Opteron 846 - 10502 (103.62%)
Opteron 844 - 10135 (100.00%)
Xeon MP 2.0 GHz (2 MB L3) - 6700 (66.12%)
Xeon MP 1.6 GHz (1 MB L3) - 5750 (56.73%)

Opteron's home run number this time around is a 57% lead. Again with this setup, Opteron could get by with PC2100 memory.

SPECjbb2000 Performance 4P Servers
Opteron 848 - 113473 (105.28%)
Opteron 846 - 107786 (100.00%)
Xeon MP 2.8 GHz (2 MB L3) - 94405 (87.59%)

Opteron's lead is held back to a still very respectable 20%.

MMB2 (Windows) 4P Servers
Xeon MP 2.8 GHz (2 MB L3) - 16528 (106.49%)
Opteron 844 - 15520 (100.00%)
Xeon MP 2.0 GHz (2 MB L3) - 13200 (85.05%)

[These two paragraphs now change from the original article].

When you compare Opteron's SPEC performance lead over Xeon - 19%, 20%, 57%, 75%, and 107% in five different quad disciplines, and let's not forget Opteron's single/dual SPEC leadership as well - it shows breathtaking arrogance from Dell and HP when it is not giving its customers the option to buy the Opteron platform, especially when IBM is doing so and Sun will do the same. To further illustrate my point, imagine the following scenario:

If Opteron's performance lead was transposed into the car world, how do you think the Opteron brand would be received? Imagine two independent car dealerships that compete against each other. They face each other across the high street, so customers play them off to garner the best possible deal. Both companies have been successfully selling the Xeon brand, but one of them decides it's time to give their customers the option to buy an Opteron. When customers are told that for the same price Opteron betters Xeon in the crucial areas that matter - 19% lower emissions, 20% faster top speed, 57% better fuel economy, 75% quicker 0-60, 107% shorter braking distance, and then find out that it can also pull twice the load of Xeon, which car do you think is going to prove to be the better seller?