AMD or INTEL

[jfri 42]

When choosing a new system I have reached two alternativesAMD 64 X2 6000+ 3 GHzIntel Intel Core 2 Duo E7200 2.53GHz 3MB FSB1066 BoxedWhich of these are most powerful? How big is the difference? Thw question refers to no overclocking.I suppose the Intel 8400 is significantly more powerful but how big is the difference? How much difference would these alternatives mean in FS9 and FSX?

[newtie 533]

I'm running the AMD 64 X2 6000+ 3 GHz with a 5% OC and it runs great. I've had no problems whatsoever.Mark

[Reset XPDR 1,271]

Check these benchmark links out:http://www.behardware.com/articles/657-6/a...on-64-6000.html (FSX RTM)http://www.overclockersclub.com/reviews/amd_6000/13.htm (likely FSX SP1)http://www.simhq.com/_technology2/technology_106c.html (FSX RTM again)and http://www.extremetech.com/article2/0,2845,2112088,00.asp (FSX RTM)Although most of these benchmarks are done on the release version of FSX (ie. RTM), it is likely the results scale the same to SP1 and SP2. Given each of these review concluded that an E6700, which runs at 2.66GHz, was roughly 10-20% quicker than the 6000+, I'd say the E7200, at 2.53GHz and of newer architecture, would be roughly 10-15% quicker than the 6000+.Gary

[Guest D17S]

That AMD Athlon >< Intel Core2 performance difference displays (only) Intel's data path optimization strategy. The Nehalem is just the same old Athlon/Core2 short(er)-pipeline CPU, so Nehalem's performance boost (we are hoping for) will be based on Intel's ability to squeeze another X% via even more optimizations. The real performance boost will Still be about increasing the number of cores. The Real breakthrough was the AMD's Athlon. We'll likely not see that again . . . and most certainly Not with Nehalem. This will be interesting, but I'm not real hopeful. I'm guessing 20% will be about it except for some very special applications (x-FSXyz). In any case, Intel is in the lead and it appears it will stay that way for the forseeable future.

[GHarrall 494]

>That AMD Athlon >< Intel Core2 performance difference>displays (only) Intel's data path optimization strategy. The>Nehalem is just the same old Athlon/Core2 short(er)-pipeline>CPU, so Nehalem's performance boost (we are hoping for) will>be based on Intel's ability to squeeze another X% via even>more optimizations. The real performance boost will Still be>about increasing the number of cores. >>The Real breakthrough was the AMD's Athlon. We'll likely not>see that again . . . and most certainly Not with Nehalem. This>will be interesting, but I'm not real hopeful. I'm guessing>20% will be about it except for some very special>applications (x-FSXyz). >>In any case, Intel is in the lead and it appears it will stay>that way for the forseeable future. With respect, the Nehalem is much more than 'the same old Core 2'The major feature for Nehalem is the integrated on die memory controller that does away with the need for the FSB. AMD have been doing this for a long time with the A64 but they have never been able to make good use of the extra memory bandwidth that became avaialable. It also marks a return for Hyperthreading albeit in a different guise.There is a very detailed Nehalem preview article at AnandTech. Well worth a read.Bottom line, with a crocked motherboard and chipset and limited to 2.66 GHz, the Nehalem was still 20% faster than Core 2 Duo at 3+ Ghz and very near to the benchmark score from the 'Skulltrail' platform.At 3Ghz+ the Nehalem should blow everything else away.

[jfri 42]

>I'm running the AMD 64 X2 6000+ 3 GHz with a 5% OC and it>runs great. I've had no problems whatsoever.>Are you reffering to both FSX and FS9? Is it with addons? What graphics card do you use?

[jfri 42]

>Check these benchmark links out:>>http://www.behardware.com/articles/657-6/a...on-64-6000.html>(FSX RTM)>>http://www.overclockersclub.com/reviews/amd_6000/13.htm>(likely FSX SP1)>>http://www.simhq.com/_technology2/technology_106c.html (FSX>RTM again)>>and http://www.extremetech.com/article2/0,2845,2112088,00.asp>(FSX RTM)>>Although most of these benchmarks are done on the release>version of FSX (ie. RTM), it is likely the results scale the>same to SP1 and SP2. Given each of these review concluded>that an E6700, which runs at 2.66GHz, was roughly 10-20%>quicker than the 6000+, I'd say the E7200, at 2.53GHz and of>newer architecture, would be roughly 10-15% quicker than the>6000+.>I found a 7200 test and it seem to confirm that 7200 is indeed somewhat better than 6000+ and the 8200 somewhat better than 7200.But they didn't test it with FSX. I also saw that in the 7200 the virtualization technology is disabled. What could that mean for FS?

[Guest D17S]

>At 3Ghz+ the Nehalem should blow everything else away.We've been through the memory bandwidth discussion and found the results to be subjective, at best. Moving the chipset onto the die is NoT a significant performance eNhancer, it IS (however) a significant production/profitability eNhancer.Nehalem's hperthreading will only help multithreaded apps. If FS11 can optimize for multicore, my current Q6600 quad will see a quadruple'd performance. Why then, should I spend spend thousands on Nehalem from a current core2/quad rig (especially when the 32nm shrink, then Sandy Bridge will be just a tick-tock away)? Let's spool back and think about this just a little more. We have to be careful about benchies v FS. Our FSX is "Special" (Remember we saw similar claims about Penryn). The 20% is about what I expect over the current 64nm Core2. Maybe 5% from the quickpath production enhancement and the rest from data path optimizations. The biggest deal is that this new buss architecture will chop $20-80 per system set (CPU+ northbridge) from the cost of production. If they don't pass that on to us, Buy INTC. That could provide a Very nice ROIC.But still, stay hopeful. We just don't know.

[newtie 533]

I'm running FSX (Vista32)with the usual suspects..UT,UTX,ASX,Flight Environment, etc. What made a huge difference for me was the use of Ken Salter's AlacrityPC. Here's an example: I load up FSX, ASX and Eaglesoft's Citation X (and everything else) at my local airport, KSTS, all sliders middle of the road. In windowed mode prior to alt + enter into fullscreen, I'm getting 19.5 fps on the runway and micro stutters in the turns after takeoff.Same exact flight running Ken's program with no changes, I get 30 fps on the runway and no stutters whatsoever in the turns after takeoff. if my 8th grade math serves me, that's a 33% increase in frame rates. I don't know enough about FSX to explain why, but I sure can see it! I think this program is of tremendous value and could make a middle of the road processor feel like something a lot more expensive.Mark