While i find the discussion insightful, it still doesn't help me. I am not into heavy overclocks, and unless the TDP drastically affects my choice of Mobo / PSU, a difference of 40-50 W does not bother me.
Sorry for going off-topic. As I say, whatever CPU you buy will perform as it should for its price point. Each company prices their processors based on where they see it fitting in, in terms of performance. A dual core Intel i3 costs 7k, a "six core" FX 6300 costs 7k - AMD is not being generous here, the reason they priced it similarly is because you will find both perform fairly similarly. If your CPU budget is 10k and you get an FX 8xxx, there's absolutely nothing wrong.
My post was only to point out the erroneous claims by john who claims that :-
a) FX processors have hardly any power difference vs Haswell <-- his 25~30W claims still nowhere to be seen, even after we've given him 4 reviews with 80W+ differences
b) Haswells use more power than FX when overclocked <-- still not proven, john's only response was a sarcastic comment about price
If John quit being defiant about his incorrect claims, we could've closed up this off-topic discussion many posts ago.
It would be great if you could give your views on :-
1. Multi Core vs Single Core performance, with an eye on the future (considering the fact that both MS and SOny have incorporated AMD multi-core solutions in their consoles)
2. Choice of mobo for AM3+ (I'm thinking 990FX - but I got a little apprehensive reading the bit about MSI mobos burning out)
1. The AMD "Jaguar" CPU architecture used in the new PS4/Xbox One consoles is an Intel Atom equivalent, i.e. nowhere near processing performance of AMD's FX processors leave alone Intel's Core procs. While its not a big deal considering consoles are optimised platforms, I guess the lesson from this is your GPU matters more than your CPU.
2. You should be fine since people in this thread haven't recommended MSI boards. You can cross-check with the motherboard VRM database :
http://www.overclock.net/t/946407/amd-motherboards-vrm-info-database - pick a board that has 140W TDP and as many phases as you can fit into your budget
Eddy kindly stop being crazy ! Re-read the post.
Please understand that :
1. We are discussing
Power consumption ONLY
2. Prime95 is used for Power consumption load tests in both your legitreviews test and bit-tech test. The
"in house" media benchmark* is NOT used for power consumption tests.
3. Prime95 results cannot be directly compared between sites
because they are using different test setups altogether. I am unable to see what the legitreviews Intel test setup is since they have only listed the AMD test rig.
4.
Even if prime95 is optimised for Intel and runs 10 times faster, it will still load the Intel chip to ~100% which will still help us evaluate its max power consumption
* The bit-tech "in house" media benchmark uses Handbrake for video encoding. Handbrake is a front-end for the x264 encoder.
Legitreviews uses an x264HD benchmark, which uses the x264 encoder again.
Please point out the intel bias in the above test scenario? If you're saying the input video files to the x264 encoder is "optimised for Intel", you are smoking some killer stuff.
P.s.: We are all still waiting for the dozens of reviews or even one review showing 25~30W difference. Is it usual for AMD fanbois to throw about exaggerated claims and then call others crazy when they can't back their claims?
