[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

advice on choosing an AMD chipset



Dear Folks,

I am planning to order the components for a new PC that I will put together myself with a bit of help from the local PC store I use for repairs etc.

At first I had thought to buy an FX 8350 motherboard plus fan and some RAM and a power supply e.g. Corsair Builder Series CXM 750W Modular 80 PLUS.

The case could be a large one e.g. Cooler Master HAF X USB 3.0 XL ATX Case.  Perhaps that is too big but it would have good ventilation.

I would add a flash drive (SSD) and an optical drive and run with it for a number of months and then get a graphics card etc - probably an Nvidia one.

I started looking at the AMD Kaveri A10-7850K processor and reviews comparing it with the FX8350 etc.

At first it seemed that the FX8350 was faster per $ or £ than the Kaveri and that a separate graphics card like an Nvidia one was a better option according to various web sites I visited on the subject.

But then I read about HSA  (Heterogeneous System Architecture) and discovered that it had been used to make e.g. libre office run a lot more efficiently.

I don't do gaming so I don't need powerful graphics acceleration.  I began to realise that the graphics acceleration produced by the Kaveri processor would probably be adequate for my own requirements.  

From what I can see looking at some tests done by Phoronix it works a lot better than typical on board graphics would do on a conventional CPU set up.

It also seems that the open source drivers work pretty well for Radeon graphics cards.

My question to you is this: if I would choose the Kaveri processor (and probably not bother adding either an Nvidia card or an extra Radeon card) and run with it, how rapidly (e.g. 2- 3 years) do you think HSA enhancements and optimisations of popular packages used in Debian etc be created and incorporated into new releases (e.g. Jessie and beyond)?

How much potential do you think HSA has?

Regards

Michael Fothergill

































Reply to: