but he does need to that's the whole point! His 'point' is that in the 2500k vs AMD 83xx benchmarks that the tech press were wrong to assume that running games at low resolutions was a good way to identify future performance by 'isolating' cpu processing power. Now he's right with that historical comparison but this is mostly due to the suite of games being used making better use of an increased core/thread count allowing the AMD chips to 'catch up'. He expects this trend to continue going forward I.e a CPU with slower core speed/ lower IPC but more cores/threads will show better future performance then a faster but less parallel CPU. On this predidiciton i think there is good reason to think that he is wrong due to physical constraints that mean that there is a limit to how 'parallel' most software can actually be and due to the latency issues that communication between higher cores counts causes. With 8c/16t CPU's we are very likely close to if not, in some scenarios, past the theoretical optimal core/thread count for most consumer workloads.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Upgrade. I'm so confused.
- Thread starter gunnerzz1008
- Start date
but he does need to that's the whole point! His 'point' is that in the 2500k vs AMD 83xx benchmarks that the tech press were wrong to assume that running games at low resolutions was a good way to identify future performance by 'isolating' cpu processing power. Now he's right with that historical comparison but this is mostly due to the suite of games being used making better use of an increased core/thread count allowing the AMD chips to 'catch up'. He expects this trend to continue going forward I.e a CPU with slower core speed/ lower IPC but more cores/threads will show better future performance then a faster but less parallel CPU. On this predidiciton i think there is good reason to think that he is wrong due to physical constraints that mean that there is a limit to how 'parallel' most software can actually be and due to the latency issues that communication between higher cores counts causes. With 8c/16t CPU's we are very likely close to if not, in some scenarios, past the theoretical optimal core/thread count for most consumer workloads.
I was talking about 1) his comparison of the AMD 83xx series to the 2500k....
I.e people would have been ill advised to buy an inferior AMD CPU (83xx) at the time over a 2500k on the basis that it would 'catch up' four years down the line....
and then 2) suggested waiting to see what the incoming hex core ryzen cpu's have in the way of core speed...
I can only assume that you we're blinded by rabid dogmatism and somehow read '7700k' when I wrote 'AMD's hex core'
Last edited:
I saw the Adored TV YouTube video too and it looked clear that the older higher core Fx 83XX processors had caught up and maybe surpassed sandybridge.
Then just watched this and there is definitely something funny going on with the way different people test and get different conclusions:
https://youtu.be/76-8-4qcpPo
Then just watched this and there is definitely something funny going on with the way different people test and get different conclusions:
https://youtu.be/76-8-4qcpPo