Has someone actually tested them
Why intel+Nvidia? Intel's iGPUs are powerful enough to take on Carrizo APUs alone.
I doubt that this is true unless you are comparing to those top of the line + L4 units. Which not only cost 600 for the chip alone but far more importantly consume like 50-60W while offering their respective performance. Might as well go Intel + Nvidia it's cheaper and more efficient.
Also Intel graphics were garbage in every way the last time I used them (2013) and unless they fixed all these issues I'd rather not use them ever again.
And speaking of efficiency, NBC recently tested both intel+Nvidia and AMD+AMD versions of the HP Pavilion. Intel+Nvidia was clearly more efficient and more powerful.
Not the top model thus clocks compensate for the extra CUs
Different screen (at 17" that matters)
Gimped ram config
Different HDDs
Obviously not a fair comparison
Besides the points there are a couple of thing that leave question marks. Such as listing an average idle power cosnumption of 10W and a max an min of 10.8 and 5.3 respectivly for the Intel+Nv based one. 10W on average must be bogus but an average is over a larger time period thus it must be representative unless they dropped the ball twice and measured for 20 seconds.
If they have the same average power consumption their duration on battery should be equal. I say should be because laptops with bigger batteries and lower average power consumption gave lower run times than this notebook.
As can be seen here:
http://www.notebookcheck.com/Test-Asus-Asuspro-Essential-P751JF-Notebook.144427.0.html
This notebook has an average idle power consumption of 9.2W...
A Bigger battery of 47Wh rather than 41Wh...
... Yet can idle for 2 hours less than the HP Pavilion
What I think is going on is that they havent checked actual capacity and it does happen that batteries come from factory with 200 cycles already obviously degrading them to the 30Wh area down from 41 or 47.
What are we doing now? Waiting for other notebooks, hoping the results will be different and favouring AMD for some reason?
Interesting question asked as if anyone is so biased. The thing is the main reason why I buy a laptop before decididing whether I will keep it or not is because I don't trust reviews. Because usually the reviewers are obviously either incompetent or payed off. Only on a rare occasion a review isn't riddled with flaws and contradictions as you can see above.
To conclude I'm probably going to try and get my hand on a fx-8800p based machine to judge for myself. We know Carrizo is better than Intel's offerings if Tom'sHw is to be believed as they claim their hand on experience has convinced them that Carrizo could be a game changer not just for AMD but for the mobile market. Additionally we have Stillt providing numbers of what Carrizo really can do and yes at 35-42W it actually matches the 840M in GTA V.
Going back to the real problem at hand with the cheap gimped a10-8700p based HP pavilion.
"Our gaming tests confirm the evaluation that the dual-graphics does not bring any advantages for the user right now. Except for a few titles like
Crysis 3 or
Tomb Raider, the
Radeon R8 M365DX is not faster than a single
Radeon R7 M360; the majority of titles show the same or even lower frame rates. You also get very uneven frame times or micro stutters, respectively, because of the
asymmetrical Crossfire, so we recommend the manual deactivation of Dual Graphics in the Catalyst Control Center and that you only use the dedicated GPU instead. Still, the Pavilion 17 only manages current titles with low or medium details and 1024x768 or 1366x768 pixels."
Using dual graphics in your product while it sucks is bad. Not sure what they are talking about this is symmetrical crossfire both chips have a matching architecture (tonga) and the same number of CUs.