A little off topic to this thread but i always found it silly that intel cuts down or underclocks their igpu's in the lower end models even though theyre already the bottleneck, other then yeilds is there a reason for this?

A little off topic to this thread but i always found it silly that intel cuts down or underclocks their igpu's in the lower end models even though theyre already the bottleneck, other then yeilds is there a reason for this?

None that I can see other than TDP. They should take a note from AMD's APU's and couple their lower end CPU cores with a beefy iGPU. Then when DX12 hits people will get a nice performance increase on chips like these as low level hardware access (which DX12 should bring) will drastically help improve performance for people with lower end CPU's and decent iGPU's. AMD's mantle has proven this that if you have a low end CPU you will be see more of a benefit from low level access. Would make sense, but then it might also put people off buying their higher end mainstream platforms and just going for the lower end hardware to save money when companies like Intel want you to buy their more expensive stuff.

A little off topic to this thread but i always found it silly that intel cuts down or underclocks their igpu's in the lower end models even though theyre already the bottleneck, other then yeilds is there a reason for this?

I asked me the same, but doesn't make sense, at least they should have some entry level CPU with the full iGPU, like G3258 with Iris Pro 5200.

What architecture's it built off? Atom or Haswell or something else?

What architecture's it built off? Atom or Haswell or something else?

Its broadwell, these are intels first broadwell chips. There more for large tablets and ultra books I believe, and they should be good for fanless ultra thin devices thanks to the TDP and small physical dimensions of the chip itself