formulav8
Diamond Member
- Sep 18, 2000
- 7,004
- 522
- 126
Or you trully belived that bs that devs need to baby sit threads?
If you mean try to use the most important threads in the best possible way for the CPU, sure why not?
Or you trully belived that bs that devs need to baby sit threads?
Problems caused by Ryzen design and the only real way to fix it is to only use half the cpu for a gaming process and its subprocess. Or you trully belived that bs that devs need to baby sit threads? Look at that F1 2016 fail because is using that obsolete design.
It might perfectly be netcode or some bugs at play. But its fun.Yes people have been reporting more fluid, smoother game play on Ryzen all along. Which makes it more curious than ever that sites like PCPer, the champions of smooth game play, refuse to investigate.
You completely misunderstood the F1 2016 issue. The issue is that F1 2016 is that when it is is installed it saves the CPU topology in the settings, the settings is backed up in steam and if you reinstall on a new system the settings will downloaded again with the topology for the old CPU.Problems caused by Ryzen design and the only real way to fix it is to only use half the cpu for a gaming process and its subprocess. Or you trully belived that bs that devs need to baby sit threads? Look at that F1 2016 fail because is using that obsolete design.
What's wrong with it is the vast majority of games (talking close to 100%, not simply over 50%), don't do their own User Mode Scheduling.She acknowledged there were games that are not playing well on Ryzen and that there are developers committed to resolve what they can with updates in the future. That was blaming developers? She simply stated there are games that aren't optimized for the new CPU. What's wrong with saying that?
computerbase.de revisits Ryzen gaming performance using Windows 10 with various power profiles and HPET On/Off versus Windows 7. Apparently with the High Performance profile enabled in Power Settings it bridges the gap with the 7700K:
http://imgur.com/a/TRaUM
Source: https://www.computerbase.de/2017-03/ryz ... e-parking/
Hopefully this will put to rest the idea that Ryzen "sucks for gaming."
Is it bridging the gap when CB.de at launch day had gap at 5% in the first place? Add that 5% to every other review result and results will still be not pretty.computerbase.de revisits Ryzen gaming performance using Windows 10 with various power profiles and HPET On/Off versus Windows 7. Apparently with the High Performance profile enabled in Power Settings it bridges the gap with the 7700K:
Their 7700k results are oddly low for many games.computerbase.de revisits Ryzen gaming performance using Windows 10 with various power profiles and HPET On/Off versus Windows 7. Apparently with the High Performance profile enabled in Power Settings it bridges the gap with the 7700K:
http://imgur.com/a/TRaUM
Source: https://www.computerbase.de/2017-03/ryz ... e-parking/
Hopefully this will put to rest the idea that Ryzen "sucks for gaming."
That BF1 bench is quite telling. I am sure there is no "bug" in the windows 10 scheduler but i'd sure like to know why win7 pulls ahead some times
Yeah, their review of cores before Ryzen launch fired up the "Ryzen will totally be good at gaming, because more cores are now all the rage in gaming!" gearTheir 7700k results are oddly low for many games.
Looncraz's idea of nV driver simply acting up on Win10 is satisfying enough.That BF1 bench is quite telling. I am sure there is no "bug" in the windows 10 scheduler but i'd sure like to know why win7 pulls ahead some times
Looncraz's idea of nV driver simply acting up on Win10 is satisfying enough.
Is it bridging the gap when CB.de at launch day had gap at 5% in the first place? Add that 5% to every other review result and results will still be not pretty.
I am sure there is no "bug" in the windows 10 scheduler
The thing that could possibly improve certain area's is optimizing the scheduler for the ccx's. But the current scheduler is aware of core/smt on Ryzen already.
No, i got it right, if the game needs to identify anything means its not using the new model of "just create threads and forget about them", creating threads and letting the O.S. scheduler do its job, thats the new and current model. If you do that its the same if is 4, 8, 16 or 128 cores, or if it runs on Window,s Linux, Android or console, your code should aways run right, as long you dont run intro main thread bottleneck. If you need to idenfity CPU then you are doing it the old way.You are misunderstanding the situation. F1 was identifying it as 16 cores not 8 cores, that's all.
In some reviews, it was likely identified as a quad ( and so were the Intels with more cores) but few tested F1 anyway.
For every CPU out there + future cpus? on every platform you game will be running?! good luck!If you mean try to use the most important threads in the best possible way for the CPU, sure why not?
What she means "if the game doest run right, its the developer fault for not optimising for Ryzen", so every game that doest run right is the developer fault because is not optimised for it... its the developer, its not the CCX design. That what that means, and it seems to have worked right because ppl are blaming the software for performance problems with Ryzen.There's a big difference between "developers fault" and optimizing for Ryzen. The basic scheduler is working but there are ways to take advantage of the Ryzen design specifically that future titles could see, or changes to current titles if the developer chooses to do so. And that's the line AMD has been saying all along, that there can be optimizations that can be performed, she's not blaming developers as if they did something wrong.
Nobody is expecting games to implement user mode scheduling, the potential 3-7% performance improvement is simply not worth it.What's wrong with it is the vast majority of games (talking close to 100%, not simply over 50%), don't do their own User Mode Scheduling.
Expecting these tens of thousands of games to suddenly get patched to implement that, many of which are made on an old engine they didn't make, is freaking absurd. Games tend to have uneven thread loads so they let the more omnipotent OS scheduler put threads where it thinks they should go instead.
This is a problem in the Windows 10 scheduler. It should prioritize putting software threads of the same application on a hardware thread that is shared with the program that is spawning it when possible, unless otherwise instructed to assign it elsewhere.
Windows is trying to aggressively to balance load among warm threads for power saving instead of trying to keep the threads of a program under the same L3. This is a "bug" in my eyes.
There are ways to further improve performance by having groups of threads utilizing each L3 cache, yes. These are the situations that go beyond what can be expected from an OS scheduler. But that is meaningless for 99%+ of current games. THAT is the case where you'd expect a game to use User Mode Scheduling, not simply to get the OS to keep all/most threads on one CCX!
No, i got it right, if the game needs to identify anything means its not using the new model of "just create threads and forget about them", creating threads and letting the O.S. scheduler do its job, thats the new and current model. If you do that its the same if is 4, 8, 16 or 128 cores, or if it runs on Window,s Linux, Android or console, your code should aways run right, as long you dont run intro main thread bottleneck. If you need to idenfity CPU then you are doing it the old way.
.
What are you smoking?
The game takes the core and thread count and it does so properly for Intel but not for Ryzen.
You need to take a step back and think before going on a raging rant. Unless your aim to troll and then, have fun.
The games does not need to read ANYTHING AT ALL. Just create the threads you need and keep on going, is the O.S. job to do load balancing and thread affinity.
If it needs to read core count thats because it runs on a outdated programing model.
If you have any doubt open the resource manager and take a closer look at the number of threads a game starts howdays.
I was going to buy 1700 for gaming only and now Im not sure with this... https://community.amd.com/community/gaming/blog/2017/03/13/amd-ryzen-community-update?sf62107357=1 ... 7700K looks like the way sadly :/
This is very simple, a program should not act diferently if it has 2, 4 or 256 cores, it is as simple as that, a modern MT code will scale properly, and i think you will have a hard time finding the "first check how many cores you have at runtime" on any programing guide. F1 2016 is using old style programing if it needs to know how many cores it has avalible. /periodROFL