Basically - you're comparing a 64-bit with a 32-bit in encoding. I may have said it was better but in all basicness I'm quite drunk now so I'm not thinking straight.
Ok, we all know what a 64-bit chip consists of and the difference between it and a 32-bit Processor. Most programs cannot do anything with the 64-bit but Encoding is different. The 64-bit can handle Millions of Numbers More, I think its about a Billion times more numbers. Encoding uses numbers entirely and if a Processor can handle more numbers no matter how much less powerful it is it is going to be able to do that better.
Now - for a 32-bit Processor to be only a fraction of a second less than a 64-bit if completely brilliant. Its an insane thing to be able to do. Even seconds behind is damn good going. The AMD-FX-53 without the 64-bit extensions would take 10 to 20 seconds more time to do any of these encoding Programs. I'm amazed the Intel Beat it once. You can see the difference with the non-64-bit AMD Processors at the bottom, 20 seconds behind even its price alternative, let alone its basic alternative.
If you want encoding - the 64-bit is in there I must admit. At the moment that means AMD but to see a 32-bit practically equal a 64-bit in handling Numbers. Respect must be shown do you not agree? And the price difference between the Fx-53 and the HT EE is not actually that much