![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
Not sure why the last thread was closed... somebody obviously has their panties in a bunch. I figured it should be continued. It was a good discussion unless the powers that be, in all their wisdom 
, have decided that debate isn't allowed here.
My response to the last post in the other thread:
Actually... this round Intel has the TDP war won. I'm building a system with dual 40watt CPUs, which will draw less power than your single FX chip. Also you have to understand that the temp numbers that people are posting CANNOT be compared to older/different CPUs since Conroe/Woodcrest/Merom/Kentsfield take their temp readings from a different location that is more accurate and will appear to be higher.
Well... considering most people don't really consider the type of memory beyond speed, ECC/non-ECC wouldn't really factor into their opinion. And I'm right about the errors, though the figures may be off a bit. The reality is that you WILL get a certain number of errors per # of writes per capacity. Now... will they be critical errors? Maybe not... but IMHO 1 error is too many when you are working on important documents/files.
Show me 1 single person running SLI 8800s that is having issues with any game at any resolution with any size monitor. Guess what.. there isn't one. Now... i know games are going to get more demanding, but then again... graphics cards are going to get better as well. Do you really think that anybody, outside of benchmark fanatics, are going to run quad cards? The only way that I can see people doing quad SLI is if they run "lower end" cards to get the performance of 2 higher end cards. Hell... you've got people complaining right now that running dual SLI they don't have enough space on the board for their other PCI cards... what do you think is going to happen with quad SLI?
My comment about the benchmarks was more to make a comment before the benchmarks come out. With server benchmakrs you will see the benefit of AMDs IMC design compared to Intels. The gains that Intel has made in server market haven't been as good because of this. Now... my point is that 4x4 is marketed towards desktop users, competing against Intel's desktop line, and to use benchmarks that are server based would be misleading. It's already and apples to oranges comparison... let's not make it an apples to steak comparison.
, have decided that debate isn't allowed here.My response to the last post in the other thread:
Astrogiblet said:
Actually... this round Intel has the TDP war won. I'm building a system with dual 40watt CPUs, which will draw less power than your single FX chip. Also you have to understand that the temp numbers that people are posting CANNOT be compared to older/different CPUs since Conroe/Woodcrest/Merom/Kentsfield take their temp readings from a different location that is more accurate and will appear to be higher.
Astrogiblet said:
Well... considering most people don't really consider the type of memory beyond speed, ECC/non-ECC wouldn't really factor into their opinion. And I'm right about the errors, though the figures may be off a bit. The reality is that you WILL get a certain number of errors per # of writes per capacity. Now... will they be critical errors? Maybe not... but IMHO 1 error is too many when you are working on important documents/files.
Astrogiblet said:
Show me 1 single person running SLI 8800s that is having issues with any game at any resolution with any size monitor. Guess what.. there isn't one. Now... i know games are going to get more demanding, but then again... graphics cards are going to get better as well. Do you really think that anybody, outside of benchmark fanatics, are going to run quad cards? The only way that I can see people doing quad SLI is if they run "lower end" cards to get the performance of 2 higher end cards. Hell... you've got people complaining right now that running dual SLI they don't have enough space on the board for their other PCI cards... what do you think is going to happen with quad SLI?
Astrogiblet said:
My comment about the benchmarks was more to make a comment before the benchmarks come out. With server benchmakrs you will see the benefit of AMDs IMC design compared to Intels. The gains that Intel has made in server market haven't been as good because of this. Now... my point is that 4x4 is marketed towards desktop users, competing against Intel's desktop line, and to use benchmarks that are server based would be misleading. It's already and apples to oranges comparison... let's not make it an apples to steak comparison.