Poor AMD... Always substandard. There's no way an FX processor can compete with a GTX 480 SLI system in terms of heating up a room.You too use it as a heater I see?
Poor AMD... Always substandard. There's no way an FX processor can compete with a GTX 480 SLI system in terms of heating up a room.
Funny how so many people thought that system was good for bragging rights, especially paired with a max overclock 45nm chip.
I already answered once here, but I'd like to revise my thoughts a bit.
Zen will be attractive to me only if single-threaded performance matches or improves upon what Intel has at a similar price, which I really don't expect to happen.
Food for thought though, if AMD can get ~Broadwell level IPC, similar performance per watt, similar clockspeeds, can get their die size down by shedding some things like iGPU and the transistors related to AVX2, and is happy with lower margins, I would totally buy a Zen quad for much less than a Skylake quad. If they can build a quad core that's half the size of Skylake (which shouldn't be too hard, given how much of SL is iGPU) and sell it for 50-75% the price, AMD can have my money.
I would hope it would be. If a Zen quad is no cheaper than an i5, with less IPC, performance per watt and clockspeed and lacking Intel's iGPU and specialized instructions, I can't see a single attractive thing about it.
No matter the year, no matter the API,FPS is locked at how fast you are able to run the main thread,if the heaviest workload you do is gaming 4 cores with higher single threaded speed will still be much better.So if they will offer Ivy/Haswell IPC with 50% more Cores/Threads at the same price you will go for Higher IPC and less cores/Treads in 2017 and DX-12/Vulkan era ????
No matter the year, no matter the API,FPS is locked at how fast you are able to run the main thread,if the heaviest workload you do is gaming 4 cores with higher single threaded speed will still be much better.
http://www.tekgrains.com/best-features-directx-12-vulkan-explained/
You could go to extreme ends to do that but it would not be beneficial,you would have to make tough tasks like enemy AI or physics much simpler and easier tasks much heavier which would not result in a good game.According to that slide, the game code is unevenly distributed across cores. Is it not possible to design game code that utilizes cores much more evenly than depicted here?
Any program with event loop will have uneven distribution of load between threads.According to that slide, the game code is unevenly distributed across cores. Is it not possible to design game code that utilizes cores much more evenly than depicted here?
The easy way of making a game engine multithreaded is splitting things like AI and physics into a thread each, but that's not the only way, or the best way of doing it. It's quite possible to parallelize the game into multiple threads that share AI, physics etc between them. That will distribute the load more evenly, and scale to more cores. Games aren't like that now because of the effort required, but that is slowly changing.You could go to extreme ends to do that but it would not be beneficial,you would have to make tough tasks like enemy AI or physics much simpler and easier tasks much heavier which would not result in a good game.
Basically almost every console game does just that, that's why so many games are awesome graphics with nothing much else.
Out of pure curiosity in relation to the required flexibility of the new machine, as of this moment, would your daughter need a more powerful CPU than an Ivy Bridge i5 @ 4Ghz+ ?Like Russian Sensation, I have recently built 2-6700k rigs, one for me and one for my daughter and son-in-law. I could have saved some $ going with 6600ks but now having experience with 2500k/3770k and 4790k the price difference from 6600k to 6700k was small enough that I opted for the more expensive cpu to allow more flexible and powerful machines.
So if they will offer Ivy/Haswell IPC with 50% more Cores/Threads at the same price you will go for Higher IPC and less cores/Treads in 2017 and DX-12/Vulkan era ????
Out of pure curiosity in relation to the required flexibility of the new machine, as of this moment, would your daughter need a more powerful CPU than an Ivy Bridge i5 @ 4Ghz+ ?
While you could be right, please note that 8350 is Piledriver, not Excavator nor Steamroller and also one application, possibly very badly optimized, is not a general trend.
No matter the year, no matter the API,FPS is locked at how fast you are able to run the main thread,if the heaviest workload you do is gaming 4 cores with higher single threaded speed will still be much better.
http://www.tekgrains.com/best-features-directx-12-vulkan-explained/
Here is a well-threaded AAA title:
http://gamegpu.com/action-/-fps-/-tps/assassin-s-creed-unity-test-gpu.html
3.6Ghz 8150 = 48 FPS
4.0Ghz 8350 = 57 FPS
4.7Ghz 9590 = 67 FPS
Vs.
3.5Ghz Haswell i3 4330 = 69 FPS
3.0Ghz i7 5960X = 107 FPS
Seems Intel is about 90-100% ahead in IPC. Excavator hardly moved IPC over Vishera.
Although, you are right that how well the game is optimized can have a tremendous impact. The successor to Unity performs miles faster on FX series.
That's why AMD really needs to work closer with AAA game developers so that Zen can shine under DX12/Vulkan.