AMD A10-6800K Richland APU On Linux
Phoronix: AMD A10-6800K Richland APU On Linux
Earlier this month AMD unveiled their Richland desktop APUs as an upgraded version of Trinity. While still based upon Piledriver CPU cores, the AMD A10-6800K APU under Linux is a modest upgrade until the arrival of the Jaguar-based APUs. For starting off our Linux testing of the A10-6800K are Ubuntu Linux benchmarks of this high-end Richland APU compared against the A10-5800K Trinity APU.
Originally Posted by chris200x9
Reading is hard.
Originally Posted by Phoronix
It really is.
Originally Posted by d2kx
AMD jucked up some of the code for it's newer CPU's
"The Overclocker "The Stilt" from Finland has managed to optimize SuperPI performance on AMD systems"
is it fixxed on Linux?
Superpi is worthless as a benchmark. It's almost entirely x87 code. It's easy to verify, just run a profiler on it.
I'm actually surprised about the difference between Richland and Trinity. I didn't think it was going to be that large of a difference. By the time I'd need to get a Richland system I'm sure Kaveri will be out though, so I think that's going to be what my next upgrade will end up being. Depends on the GCN situation on Linux at the time, however.
ok... summing up:
if AMD goes to smaller feature sizes and further increses frequency, it will be a hell of a CPU... i mean 65% increased performance per watt is quite a "hausnummer" Germans would say
what does it actually mean that the time graphs go further.. does it really take longer? i somehow get puzzled there everytime. as it looks like the higher frequency with slightly higher power consumption would need longer. which for average may do no harm but for total consumption per task be a total disaster and would be inconsistent with the performance numbers from the other graphs.
plase, can someone explain these graphs to me? (FLAC audio encoding is a rather remarkable one of that kind...)
" until the arrival of the Jaguar-based APUs." Ahem Jaguar based apus are for mobile platform(kabini), and not upgrade for Richland. That update is called kaveri and cpu cores are called steamroller.
Originally Posted by phoronix
I'm not even suprised anymore. It's a classical Phoronix benchmark
If it's an nVidia benchmark, there are no ATi Cards, if it's an Intel
integrated graphics, they're not compared to ATi's APU's, and if it's
an AMD CPU you can be damn sure there won't be any Intel processor
thrown in for a good measure.
What's the point of this? If there is any inherent value in benchmarking
it's so that customers can evaluate performance/price/wattage ratios.
Why is it so hard for Phoronix to get this right is beyond me. Pshaw!