Trying to fight against a Bay Trail and failing hard on CPU, on the other side it brutalizes Bt on GPU. That's how I see that benchmarks.
Cant figure that one out. I would never trade cpu performance for gpu performance in that range, which is what Intel basically did.
They are not clocked the same either for the SKUs used. Else they would perform identical.
The N3050 looks ideal for a NAS. 14W idle, 17W load. (With 2x4GB and 1TB MX200 SSD)
Maybe they tried to smash Kabini and failed... To make it worse, AMD is NOT updating AM1+ and still are strong as before and even worse... Kabini are OCableCant figure that one out. I would never trade cpu performance for gpu performance in that range, which is what Intel basically did.
I was looking at that one. Seems the most promising, but i was expecting better performance. Worse than sempron 3850 is not good (even at half power)
Just get the N3700 then.
Sadly even a J2900 is better CPU wise and a Athlon 5350 is GPU wise.
The only good side is the Power consumption...
If.VIA were more active... This is the perfect moment for a counter attack.
GPU doesnt matter in a NAS.
And 25W vs 6W is quite a difference. These are tablet leftovers so to say. An area out of reach for AMD and VIA.
You can just as well use 35W T Celerons/Pentiums to compare with 25W parts.
Yes,the other option would be an embedded celeron U.
Too bad i can't find any Msi eco braswell review
If idle is your main concern. Get a Haswell Celeron if you dont pick Braswell. But remember the mobo have a huge influence here.
I think the NUC would be a bit limited for NAS. I want 4 hdds.
something weird in reviews though.
Techspot Bay Trail power figures are quite different from these:http://www.technikaffe.de/anleitung-307-asrock_n3150_itx_im_test__14nm_braswell_vs._22nm_bay_trail
Both are not far from the A4-5000:http://www.tweakpc.de/hardware/tests/mainboards/asrock_am1h-itx_asrock_qc_5000-itx/s06.php
Which lets me wonder if AMD will release a similar Carrizo part in a few months, which might be interesting
Don't forget the PSU. That can have an equal impact.
You cant use cross compare reviews with different setups in terms of power consumption.
Power differences should be consistent though.
Once taken into account power efficiency one should expect roughly the same difference in power consumption.
The techkaffe review use a 65w psu, the other a 90w picopsu IIRC.these are probably more realistic figures
We're not talking about 'normal' ATX PSUs, though. We're talking about 65-90-150W picopsus, like the ones used with Intel NUCs, and all laptops. The kind that output 19V DC, like this one: http://www.newegg.com/Product/Product.aspx?Item=N82E16817104184One PSU could be 25% efficient at 10W(40W) while 50% efficient at 20W(40W). And another would be 50% at 10W(20W) and 60% at 20W(33.3W).
Don't do that! Brasswell Sadly is a brutal step back of Bay Trail. You are dooming your father to a Pentium D levels.Has Braswell reached NUCs yet? I'd be interested in picking up an N3700 or equivalent for my father, to replace his aging 1.8ghz Core2Duo system.