Problem is without any GPU, it's going to be ignored by OEMs. There is no way they can get any kind of real volume out of it.
Actually I believe they are doing the right thing. Competing on price killed AMD. They need high end parts that can bring in the dollars.
I find one thing interesting. Some of AT forum users who were very vocal about AMD's inability to compete in high end desktop and server markets are now bashing AMD's decision to get back into those markets by developing a core that can compete head on with what intel will have next year. I thought they would be happy about that decision !? I guess they are not :biggrin:Actually I believe they are doing the right thing. Competing on price killed AMD. They need high end parts that can bring in the dollars.
Problem is without any GPU, it's going to be ignored by OEMs. There is no way they can get any kind of real volume out of it.
Problem is without any GPU, it's going to be ignored by OEMs. There is no way they can get any kind of real volume out of it.
AMD has a track record of over promising and under delivering. That has upset enthusiasts, can't really blame them for being bitter.
This roadmap is very encouraging
I can see why you might be encouraged!
Your post makes no sense.Yes, the return to high performance FX processors means people won't have to settle for intel's meager couple percentage point performance increases.
Also, it's funny you now consider it a desktop release. When estimating Skylake desktop performance increase, you keep comparing to Haswell only to avoid mentioning the embarrassing 5% yearly Intel performance increase...
Compare that to AMD Zen's 40% IPC increase. And then we're not even including any frequency increase going from 28 -> 14 nm! That's like close to all the aggregated performance increase from Nahalem->Skylake, and the Intel crowd calls it a failure. How sane is that!
Intel's top Quad in mainstream clocks at 4.0Ghz, Intel's Octacore clocks at 3.0Ghz.I already mentioned that. If Intel releases both Cannonlake and Icelake in 2017, I have a hard time seeing how most OEMs would not simply skip Cannonlake. Too short sales window.
Simple: 8 core Zen will demolish anything in Intel's top end mainstream lineup in MT performance and come very close in ST performance. So in the top end desktop segment it'll be a winner.
BD was just way ahead of it's time & wasn't refined enough to do the tasks that it was supposed to, all within a reasonable TDP.
Your post makes no sense.
The FX processors will perform worse than Intel's when they are first launched and their is no guarantee of impressive performance increases from AMD in the future, as this much hoped for 40% is a one off, due to the terrible IPC of the current FX line.
Do you think that Zen is going to clock higher than Piledriver? D:
The fact that they have switched from a speedracer design to one that has much higher IPC, would suggest that is very unlikely.
That s not 40% over the FX line but 40% over an Excavator core, this latter should have 12% better ST IPC than the FX s Piledriver core, the IPC difference with a FX core would amount to 57%.
Intel's top Quad in mainstream clocks at 4.0Ghz, Intel's Octacore clocks at 3.0Ghz.
Why is it that you think AMD is going to be able to defy the laws of physics if they release a CPU with much improved IPC?
If there is one thing that can be safely said, is that 8 core Zen, will not be coming very close to the i7 4790K's ST performance.
Is this Excavator core going to get released in any AMD product?That s not 40% over the FX line but 40% over an Excavator core, this latter should have 12% better ST IPC than the FX s Piledriver core, the IPC difference with a FX core would amount to 57%.
Yes suddenly this 40% theoretical increase based on who knows what, without a single benchmark, has become accepted fact. And possible clockspeed decreases are conveniently ignored. Typical.
FWIW they used 7 Zip, CB R15, PoVray, Blender, TrueCrypt, X264 for Kaveri s claimed 10% average IPC improvement, Blender/Povray/CB numbers in both ST and MT were used for thoses tests.
Zen sounds very promising indeed. A 40% increase in IPC over the improved Excavator core is substantial.
Without a GPU on die, it can be paired optimally for discreet GPUs that leaves more room for a higher core count and it makes sense for AMD to bring it to the enthusiast desktop market first considering their ambitions for the gaming market. The high core count will also be timed perfect for the upcoming low level APIs tailored to take advantage of the cores available to drive. Mainstream will still be addressed by high performance APUs on the common socket too, so a nice upgrade path as well.
Is this Excavator core going to get released in any AMD product?
Intel's top Quad in mainstream clocks at 4.0Ghz, Intel's Octacore clocks at 3.0Ghz.
Why is it that you think AMD is going to be able to defy the laws of physics if they release a CPU with much improved IPC?
I find one thing interesting. Some of AT forum users who were very vocal about AMD's inability to compete in high end desktop and server markets are now bashing AMD's decision to get back into those markets by developing a core that can compete head on with what intel will have next year. I thought they would be happy about that decision !? I guess they are not :biggrin:
Actually I believe they are doing the right thing. Competing on price killed AMD. They need high end parts that can bring in the dollars.
Intel's 22nm i7-5960X ships at 3.0 GHz, but it's an unlocked multiplier and most samples can go to 4.0 GHz or more. There is no particular reason to think that AMD, with access to a better process in 2016 (14nm FinFET+), could not do 8 high-IPC cores at 4.0 GHz stock if they wanted to. AMD is traditionally more aggressive about running their chips (both CPUs and GPUs) closer to the margins than Intel.
Actually I believe they are doing the right thing. Competing on price killed AMD. They need high end parts that can bring in the dollars.
Intel's 22nm i7-5960X ships at 3.0 GHz, but it's an unlocked multiplier and most samples can go to 4.0 GHz or more. There is no particular reason to think that AMD, with access to a better process in 2016 (14nm FinFET+), could not do 8 high-IPC cores at 4.0 GHz stock if they wanted to. AMD is traditionally more aggressive about running their chips (both CPUs and GPUs) closer to the margins than Intel.
Intel's 22nm i7-5960X ships at 3.0 GHz, but it's an unlocked multiplier and most samples can go to 4.0 GHz or more. There is no particular reason to think that AMD, with access to a better process in 2016 (14nm FinFET+), could not do 8 high-IPC cores at 4.0 GHz stock if they wanted to. AMD is traditionally more aggressive about running their chips (both CPUs and GPUs) closer to the margins than Intel.