安装 Steam
登录
|
语言
繁體中文（繁体中文）
日本語（日语）
한국어（韩语）
ไทย（泰语）
български（保加利亚语）
Čeština（捷克语）
Dansk（丹麦语）
Deutsch（德语）
English（英语）
Español-España（西班牙语 - 西班牙）
Español - Latinoamérica（西班牙语 - 拉丁美洲）
Ελληνικά（希腊语）
Français（法语）
Italiano（意大利语）
Bahasa Indonesia（印度尼西亚语）
Magyar（匈牙利语）
Nederlands（荷兰语）
Norsk（挪威语）
Polski（波兰语）
Português（葡萄牙语 - 葡萄牙）
Português-Brasil（葡萄牙语 - 巴西）
Română（罗马尼亚语）
Русский（俄语）
Suomi（芬兰语）
Svenska（瑞典语）
Türkçe（土耳其语）
Tiếng Việt（越南语）
Українська（乌克兰语）
报告翻译问题
讨论规则及指引
举报此帖
Put all the GTAV Graphics to the Max. Disable DoF and Motion Blur, along with MSAA.
Then go to Advanced and try setting the Up-scaling (which is actually SMAA) to around 1.25X or 1.50X
One thing to note though it is more of a "Con" to have the Grass cranked up for playing GTA Online, since it allows room for people to hide. Most will turn the Grass down for that very reason.
It would take some heft off of the CPU in games that are more GPU dependent.
1440p would balance things a bit because 1080p is a res that can really stress the CPU more; this is why many Gamers are sticking with Intel.
This
Upping the resolution would make it more GPU bound than CPU as it currently is.
Ofc, it creates much more GPU load also, but this is completely different question.
Most good benchmarks of the 2700X vs 8700K show something like a 10-25% lead for the 8700K in games and the actual fact for why an i5 8400 with a 2.8 GHz base clock but 4.0 GHz turbo and 3.8 GHz all-core turbo clock and 6 cores 6 threads may run games better than even the Ryzen 7 1800X is that Intel 1) Have lower memory latency and L3 latency between CCXs 2) 256 bit FPU and 3 which you'll also see in benchmarks if you check results for i5 8400 vs i5 8600K vs i7 8700K) Games don't necessarily need the extra threads and cores but rather become limited in performance by waiting for RAM or having 100% load on one core. Step it up to the i5 8600K and you also get a clock advantage and with the i7 8700K you get the additional 6 threads improving multi-tasking performance but gaming performance not all that much. The i5 8600K will likely beat the Ryzen 7 2700X in many more games than vice versa as is because of the advantages it have even though the Ryzen 7 2700X can run an additional 10 threads which obviously isn't compensating the higher memory latency and lower clock (especially when both are overclocked.)
With the new Ryzen processors it's also the case that the single-core clock may go up to 4.35 GHz whereas a common all core OC multiplier is 4.2 GHz which is lower and hence doesn't necessarily help in game scenarios whereas with the i7 8700K the base all core turbo is 4.3 GHz and a common all-core multiplier overclock without delidding would be like 4.9 GHz maybe 5.0 too. The Ryzen 7 2700X do have 33% more cores and threads but it has an all core turbo of 3.9-4.0 GHz and the i7 8700K overclocked to 5.0 GHz would have a 25% higher clock than that making it catch up much of the multi-core difference.
As said Ryzen can't do all that much to catch up in single-core clock though but it can catch up some with tighter memory timings but standard RAM speed on Intel is lower and maximum RAM speed on Intel is higher so you can tight things up a lot on Intel too. Though the worst case latencies are already better there from the beginning.
Anyway, considering the limited multi-plier overclock potential and value on the Ryzen if you consider the possibility to clock the i7 8700K up to 5.0 GHz maybe it's more like 20-25 to 45% faster or so in games.
The Ryzen 5 1600 have a higher memory latency than the Ryzen 5 2600 would have but what you can do to improve things is to overclock it and check the Slith (or Stilth?) or whatever RAM timings guides / programs for a guidance and possibly overclock, overvolt and tighten the timings of your RAM a bit lowering memory latency and improving the result. To some extent. But it still won't be 8700K results of course.
It's bad people do WQHD Ultra benchmark runs and such where the capacity of the CPU will be less visible because you increase the impact of the GPU possibly to the level where the CPU become completely irrelevant and they show the same result even though the CPUs aren't the same. If you want to benchmark a CPU by a game then you of course need to try to limit the impact of everything else. If you want to test the GPU instead then you should of course have the best CPU and RAM and only if you want to benchmark a specific game at a specific hardware at a specific setting you should do that. If the purpose in the later case is just to show "How are A + B running C in settings D?" Then the other stupid / common settings may serve a purpose but the extremes are much better for finding the limits of the equipment.
As for how well the games should run I don't know. I've played lots of Rise of the Triad but almost only multiplayer and it ran like crap on my CPU. The thing is though that the single played campaign ran even worse and I know Gijz have explained why and I think it supposedly possibly could be fixed by editing the maps / improving on the design there I think but that's unlikely going to happen. As for how terrible it would run even on the i7 8700K I don't know.
I don't know how bad the games run for you.
Some games had updates after Ryzen was launched and had their performance improved and I could assume that's more likely very large modern titles where they bothered with updating and tweaking for it and it's also the modern large titles which is the most likely ones to be used for benchmarks. But there's of course a lot of games which won't get updates specifically for Ryzen which will continue running worse than what could be possible. Lots of games are likely designed for quad and dual core era. There have also been OS and firmware tweaks and you can tweak a bit yourself by overclocking and running the RAM faster as said. To make it as good as possible.
I don't know how bad results you should expect.
I don't know how well you can get them.
And I don't know how good the very best chips for that task on the market would run them.
But my suggestion is to view the load numbers to see what's limiting performance and to overclock RAM, tighten timings and overclock the CPU. GamersNexus had a recent video where he talked about safe RAM voltages and also mentioned the core, SA/IO and SOC max voltage recommendations for various processors. Though for Ryzen I think AMD mentioned long term 1.35 volt but he said 1.42 but maybe that wasn't long-term. More volt is worse regardless of course. It's always worse, the question is rather how bad can you accept.
And it wouldn't improve the game performance whatsoever if the problem is the CPU. It would make frame-times a bit more equal by rising the best ones but the worst ones would remain at-least as bad.
Saying "accept lower frame rates" to someone who play fps games as a solution will unlikely be seen as much of a solution.
As for the loads I guess all of storage media, RAM and CPU can affect those.
He surely don't want to be slowed down by the CPU but he don't want to "solve" that by being even more slowed down by the GPU =P
Typical AMD Ryzen review comments: "Run the game in 4K instead!" yeah.. that solves it! ;D
Yeah, so if he want an even slower game he can do that.
But I think he wanted the games to run better not worse.
At 1080p if you are 80-90fps lots of he time but are dropping FPS below 60fps due to the GPU being starved and casuing wild fluctuations up and down inducing stuttering, where at 1440p it will eleviate the limitation and give you more steady FPS without the wild fluctuations causing stuttering and wild drops.
This advice helped lots of people over in the AC:O forum with 4 core i5's and ryzen with a 1070 or above, some cases it wouldn't be above 60fps but the gamers would be happy to just get rid of the stuttering you get with a CPU starving a GPU.
https://www.youtube.com/watch?v=mCKwPulZXgU
So it's either the gpu, your drivers, or something else.
You may get less slowdowns, but not because you rise of your the minimum fps but because you reduce your max fps.
Rising your resolution will increase the work that the GPU has to do, but won't reduce the work that the CPU has to do with your current game settings.
i7 X-Series i7 6850K 6 core 3.6 @4.4+ looped cooled max SLi lanes
SSD 2X480GB HDs + RAID O + SATA 6GB +Ram Cache
32GB RAM PC DDR4 PC 3000 GSKILL RIP JAWS why 32 was get 16 get 16 free thats why
1080p 1440p they run flawless max AA Vsync on highest settings in game some in NVCP @1080p over 100 fps low end max 144 97% of time 128-138 FPS