Makaveli
Diamond Member
- Feb 8, 2002
- 4,793
- 1,256
- 136
Oh heck yeah!
lol I was hoping for alittle more detail with possible examples
Oh heck yeah!
Adam you never answered the question tho, does it feel faster than the previous rig minus looking at benchmark results.
I call BS. There is no way anyone can tell the difference between a 1366 or 2011 system if both systems have same other components (RAM, SSD, etc...) and have a fresh installation of Windows on them.
That is seriously fast. Thanks very much
I can. I write and recompile code all day long, and I can tell you. I know how long every one of my projects takes to build.
The comment I responded to had no mention of "feel" if you go back and read it.
I knew people would start going there. High end is not part of AnandTech anymore and the brain disconnect as to why people do this often leads down these paths. I often question why I remain here. There are other forums with people that share my affliction.
+1
Most everyone here is now on a performance per $ mindset.
Hey Adam, do you play Arma series? If so, could you bench the new CPU with your GPUs in that game as there have been discussions on our forums that this game is CPU demanding. It would be interesting to see how the Gulftown compares.
Also, can you link up some pics of your CPU @ load to see what temperatures it reaches on the H80 @ 4.5-4.75ghz?
Not everyone, but a lot of people are. That's true. A lot of people have been building computers for a LONG time. For example, in hindsight, looking back, was there any difference between buying an Athlon XP1600+ vs. say XP2000+? Nope, both became obsolete just as fast. What about Pentium 4 "C" 2.4 vs. 2.6 vs. 2.8 vs. 3.0ghz? Nope. All of those topped out at around the same 3.2-3.4ghz, also became obsolete equally fast. What about Core 2 Duo E6300/E6400/E6600/E8400? Nope, all of those CPUs also becames equally obsolete despite overclocking to 3.4ghz+. Same can be said for i7 920-960, and will be said of i5 2400/2500k/2600/2700k, etc.
There is nothing wrong with a person buying a 3960 over 3930 or a 3930 over a 2600k or a 2600k over a Phenom II X4 965. It all depends on their budget, perception of value, income, and what they want out of this hobby, what programs they use, etc.
I think people who buy a $900+ processor aren't necessarily considered with "value", at least not in a traditional sense. I mean 3930 costs almost half and is only 4% slower or so. I just think with consolozation of games and most programs NOT being able to take advantage of more than 4 cores, we are bound to see people question any value in a 6-core CPU, regardless whether it is sold by AMD or Intel. That's a fair game bth. I also don't think a person buying a core i3 and overclocking is any less of an enthusiast than someone buying a $1000 Core i7-3960K. They just have different priorities, incomes (all those other points I listed). Like the argument for buying a 3960 for games only and pairing it with 2 GPUs cannot be made at all, but what if you play games with 3 GPUs and encode video at the same time?
I think a lot of people are actually interested in how much faster SB-E is in overclocked states over the older overclocked 980/990X. I think it's fair to say that "feel" wise is pretty subjective when actual performance benchmarks are widely available. In programs where 6 C/12T CPU can be used to its fullest, SB-E smokes 2500k/2600k CPUs. Also, one can clearly see that its faster than an overclocked 980/990X.
Of course there is nothing wrong with getting a 3960K even if the performance difference is only 15% over the 980X.
I don't think you should jump to conclusions that people are jealous automatically. Perhaps, they really want to know how much performance increase is necessary out of a CPU to "feel" the difference daily? Is it 10%? 15%? 20%? 25%< etc. Personally, I'd rather see actual benchmarks rather than go by "feel" since that tends to be pretty subjective.
You can tell because you base your conclusions on empiric evidence (compile time) vs 'feeling'. The original quote being: does it feel faster than the previous rig minus looking at benchmark results. Keyword here is 'feel'. If you don't look at benchmarks (in your case compile time) there is no way you can tell the difference between two extreme systems.
I apologize if this has already been answered.
I was looking at the power usage on various sites regarding SB-E Desktop 6-core CPUs, and have a theory :
It is known that this is actually an 8-core die, with two of the cores being disabled for these first two SB-E models. I wonder if they haven't been cut out of the power distribution, and although not accessible in any real way, are still using power. IIRC this is similar to the Athlon II 631, based on the Llano die, but with the GPU inaccessible. The AII 631 still has the same TDP ratings and power usage to Llano AFAIK.
It really comes down to :
(1)- The 3930 and 3960 are still using power for the two 'off' cores. Which means that the 8-core products will have similar numbers to the current products, almost no difference.
or
(2)- The 3930 and 3960 are not using any power at all for the 'off' cores. Which means that the 8-core products will have 33(?)% higher power usage than the 6-core product. This seems unlikely to me.
Bumping my question as I never got a reply.
Cliffs : SB-E with two cores disabled (from 8 to 6), does/will it use the same power as the ones with all 8 cores enabled (Xeon only?)