Because the data speak for themselves. The i5-2500K is not that much faster than the 8120 in single-threaded applications, the two trade blows through most benchmarks, and the 8120 isn't that much faster in multi-threaded apps and workflows. As I said, aside from power consumption, the 8120 is not a completely inferior CPU.
Again, as I said, if you live near a Microcenter, right now you can get an 8120 + a nicer board for $200. A 2500K + a nicer board are going to set you back at least $250-300 depending on what sales you can get. That's a cost increase of 25-50%. The 2500K is not 25-50% faster.
Many of this forum's posters need to put the Kool-aid down. If you can't interpret benchmarks, then you're just bleating like sheep.
What about the overclocking? Both chips can OC to about the same speed (I'd say 8120 a little more: 4.8 vs. 4.5 GHz) - the 2500K with it's IPC advantage and power advantage should win handly with both chips overclocked. That is something also to consider.
Because the data speak for themselves. The i5-2500K is not that much faster than the 8120 in single-threaded applications, the two trade blows through most benchmarks, and the 8120 isn't that much faster in multi-threaded apps and workflows. As I said, aside from power consumption, the 8120 is not a completely inferior CPU.
Again, as I said, if you live near a Microcenter, right now you can get an 8120 + a nicer board for $200. A 2500K + a nicer board are going to set you back at least $250-300 depending on what sales you can get. That's a cost increase of 25-50%. The 2500K is not 25-50% faster.
Many of this forum's posters need to put the Kool-aid down. If you can't interpret benchmarks, then you're just bleating like sheep.
Absolute and complete bullcrap. Did you even read the graph I just posted?
It shows that on average the 2500K is 32% faster than the 8150 in single-threaded, which has a 5% higher Turbo Core speed than the 8120. That means the 2500K is ~37% faster than the 8120 in single-threaded. This will obviously have a small variance depending on Turbo. Now, explain to me clearly how a difference of nearly 40% is "not that much".
And since multi-threaded is another part of the equation, take into consideration that the FX-8150, which has a 16% higher baseline clock speed than the FX-8120, is a mere 1% faster than the i5-2500K in multi-threaded. That means the 2500K is anywhere from 10-15% faster than the 8120 in multi-threaded depending on how Turbo is enabled for each CPU.
And if you want to mention overclocking, the situation looks worse because of the horrible efficiency of Bulldozer, especially with higher voltage.
Now make a concise effort to tell us exactly what the benefit of going with Bulldozer is, because I'm not seeing it. I'm very unbiased, and I've said it for a long time: Bulldozer is an absolute piece of crap in its current state. There's not one redeeming quality to it.
He said the 2500k needs to be 20-25% faster in order to justify the cost gap.
You point out its 10-15% faster. I don't think you are really arguing anything.
What does a user today need to do that BD can't do, but a 2500k would enable them to do? (ie, what takes it completely off the table)
I cherry picked a single non-real world benchmark to make an argument.
Because the data speak for themselves. The i5-2500K is not that much faster than the 8120 in single-threaded applications, the two trade blows through most benchmarks, and the 8120 isn't that much faster in multi-threaded apps and workflows. As I said, aside from power consumption, the 8120 is not a completely inferior CPU.
Again, as I said, if you live near a Microcenter, right now you can get an 8120 + a nicer board for $200. A 2500K + a nicer board are going to set you back at least $250-300 depending on what sales you can get. That's a cost increase of 25-50%. The 2500K is not 25-50% faster.
Many of this forum's posters need to put the Kool-aid down. If you can't interpret benchmarks, then you're just bleating like sheep.
FTFY.
That's a compilation of more than twenty applications, and the article is available here:
http://www.tomshardware.com/reviews/sandy-bridge-e-efficiency-core-i7-3960x,3075.html
Keep it up with the endless excuses.
Why is it that you have to attack everyone that is trying to be objective?
I cherry picked a single non-real world benchmark to make an argument.
The cost gap is 20-25% according to you.
There shouldn't really be any debate over this. Buy the i5 OP.
Chances are if the OP were doing anything niche that the 8120 beats the i5 in he would likely already know about it and not have to ask.
i5 has:
- Lower power consumption under load, they are both similar at idle
- Better upgrade options with ivy bridge
- Better overclocking capabilities
- Integrated GPU, this is far more useful than people give it credit for, your discrete one breaks or you sell it, then use the integrated one in the mean time With the 8120 you havent got that option
There have been a few of these bundles picked up in the DC forums, I haven't seen hard data but it sounds like the 8120 @ 4.0ghz is performing appreciably better than Thubans @ 3.5 Ghz.
Being that you have no evidence of this I guess we should all just take your word for it?
Remind me, what was it you were saying about baseless arguments again?
A majority of power users don't care about the absolute performance of their PCs or ponder their power consumption.
There shouldn't really be any debate over this. Buy the i5 OP.
Chances are if the OP were doing anything niche that the 8120 beats the i5 in he would likely already know about it and not have to ask.
i5 has:
- Lower power consumption under load, they are both similar at idle
- Better upgrade options with ivy bridge
- Better overclocking capabilities
- Integrated GPU, this is far more useful than people give it credit for, your discrete one breaks or you sell it, then use the integrated one in the mean time With the 8120 you havent got that option
Actually that is not true, there SHOULD be a debate. True, the i5-2500k outperforms the FX8100. But SLK claims it is selling for $135. So the question is not which is better if they are in different price categories. The question is about the performance of Bulldozer for it's given cost.
Perhaps I need second account so I can post one line OPs so that I can threads to read? This subforum gets a little slow some days