Which is better vid encoding Stock AMD FX-9590 vs. i7-5960X OCed to 4.2

[LeftyInSpades13]

Hello everyone, I'll try to make this short.

Every 3 years I buy a new main rig.
This year it's between a stock AMD FX-9590 running 4.7 and 5.0 turbo and a company OCed to 4.2 i7-5960X .

Now, price is not a concern. Both will be running on about the best I can find or get for them, within reason anyway.
AMD build:
AMD mobo will be an ASUS Crosshair V Formula-Z AMD 990 ATX
AMD RAM 32 GB DDR3/2133MHz Corsair Vengeance
1TB Samsung 850 PRO Series
AMD Radeon R9 295X2

Intel build
i7 mobo will be ASUS ROG X99 Rampage V Extreme ATX
i7 RAM 32 GB DDR4/2800MHz Corsair Vengeance
1TB Samsung 850 PRO Series
AMD Radeon R9 295X2

The cooling will be;
XSPC RayStorm CPU Block,Dual D5 Pump & Reservoir,AX360 Radiator with 6 fans in Push/Pull

I left out all the internal storage and other stuff except the main drive that everything will be done on, apps & warez wise. With that said, I haven't played games since the early 90s so I don't care in the slightest how great this or that is for gaming, but I am a big media enthusiast.

It seems most of the bench-marking and performance tests I keep finding for these things are game oriented or professional oriented in fields in which I have no interest or basically software isolated testing, that I don't personally understand their application to what I do.

What I would like to know, is the real life application for what I do between these two builds. Simply, between these two which will be able to better and faster handle encoding on Handbrake, couple or few browsers open with a couple hundred tabs, Administrating a website, having a media player running and something like an XMPP chat running, all at the same time?

If possible, I would like reasonable support for your recommendations or conclusions, so that maybe I can further educate myself, not the typical fanboy reply about [insert manufacturer] is bestestest.

Thnx for any help.

 

[mdocod]

@mdocod
Almost all hardware decoding is the same, for the most part. I mean as far as I'm concerned, which is in the video realm, not gaming realm, i.e. MPEG1 / MPEG2PS / MPEG2TS / AVCHD / MP4Part1 / MP4Part2 / AVI(XVID) / AVI(MotionJpeg) / MOV / WMV / MKV / WEBM / 3GPP / MP3 / WMA / LPCM / JPEG / MPO or in simpler terms, basically H/X.262, H/X.263, H/X.264. In the simplest terms, pretty much Xvid, DivX and H/X.264. The problem is, in my realm simple H/X.264 MKV has been pretty much dead for the better part of 4 years, with Hi10P, Hi444P taking over while now H/X.265 HVEC is starting to get really hot. I'm not sure if you remember or even cared, but back when H/X.264 MKV took over from Xvid AVI, it was revolutionary. The introduction of Hi10P was nice because you could get rid of a lot of compression artifacts, banding, haloing and so on much easier, but it was nothing ground breaking. H.265 though, is kinda like the jump from AVI to MKV. Only thing is, it took hardware almost a decade to get up to speed and start decoding it as a standard. I don't imagine Hi10P or Hi444P will ever be hardware decoded, but I at least have some prayers that H/X.265 won't take that long to be hardware decoded, since it was officially announced that it was going to be the new format for DVD/Bluray taking the place of H.264. Yes, software decoding is ALWAYS better than hardware decoding, the problem is to properly decode a good high-end, aggressive, Hi10P 1080p, with 5.1 FLAC you pretty much need at least a Dualcore running at 3.0 or better, which Digital Media Players generally don't have, plus the GPU can't decode Hi10P H/X.264 or H/X.265, so that's why I was wondering if there's ways to crack proprietary hardware OSs in gfx cards to introduce BETTER decoders. Plus I would like to see how one of these big ass gfx cards handles like a 1080p Hi10P or H.265 encode if it had the right decoders.
Ooops, sorry for the off topic tangent.

I'm not aware of any effort to flash/mod the existing hardware decoders to support new formats. The most likely approach would be leveraging the SIMD shader/instruction pipeline to assist in encoding/decoding. As you are already likely aware, this has been done with CUDA/openCL in some applications, the results are acceptable in some cases but most professional editing/trans-coding will still need to be done on a traditional CPU to get the best results. Point being, a big huge GPU sometimes offers a video editor very little in terms of usable assets. If you are aiming for quality outcomes, spend your money on a big 8 core haswell E and think of your GPUs as nothing more than display ports.