Ive been kinda waiting on ryzen to pop up but about a half a year ago my 2600k so had to quickly replace it.
Long story short I got a 390x and seeing its basically the same performance wise as an rx 480 I was wondering if the cpu could be a bottle neck.
Right now im running it at 4.2 ghz stable but I never really see similar performance figures close to what I see on benchmarks when it comes to rx480/390x on stock newer chips from intel
There is no way a 2700k @ 4.2 is a bottleneck for an R9 390x in most games. Keep in mind a lot of benchmarks are done in single player, on small parts of the map, in usually sub 1 minute runs. That means most benchmarks are not really good indications as to the exact fps you should expect in all parts of the game, but they can indicate which card will generally perform better and to what rough percentage degree better. Monitor your gpu utilization. I would assume that in most games your going to be having very high utilization, deep into the 90% range. If your not getting those gpu utilization numbers, then you would be able to observe a cpu bottleneck, but I find that extremely unlikely.
Thats my thoughts exactly, im kinda out of the loop with all the tools nowhere days. Anything youd recommend that I could check the utilization on ?
Bottlenecking is kind of a weird term because with every card, at some point or another, there is going to be a CPU that comes out that will take slightly better advantage of that card than the flagship CPU before it.
My question for you to help you answer your question is: does it feel fast enough for you? are you experiencing any serious lag behind other processors?
I say this as a 2600K owner - I have a Nano. I have not once felt the need to upgrade, even with Ryzen 5 just around the corner. It's still wicked fast for my needs, particularly in gaming. It beats new i5s in several gaming benchmarks I've seen, crushes the i3s (of course) and keeps up beautifully for its age with the newest i7 and Ryzen CPUs.
Now, having said that, if you need a CPU for a new workstation, I would recommend upgrading. If all you're doing is gaming, however, my opinion would be to hold out. Of course, if you want new storage mediums like NVMe and PCIe 3.0, that's cool too, and might be enough incentive to upgrade, but for my purposes, I'll likely be sticking to Sandy Bridge for quite a while.
Just my two cents, bit of a rant, all of it is opinion and subjectivity and all that, of course. tl;dr, I think if it feels fast enough for what you want, it's fine the way it is. A much quicker and potentially less expensive way to make your games run faster imo is to get a newer graphics card or add another card. That'll yield much larger performance gains.
You are seeing the same performance, because it is basically the same performance...
This is a review from a week ago... 390X is basically 2-3-4% faster than 480, and that means in many games 1-2-3 frames per second. Yes, it's basically the same performance...
Even 5% faster in most games will be 4-5fps, and you will never notice the difference between 80 and 85 fps...
So no, your CPU is not a bottleneck.
You can use MSI Afterburner to monitor gpu utilization.
redundant much redundant?
sorry...
I agree with your post except for this part:
That isn't necessarily true. GN did testing on the 2600K both @stock (yeah I know he has a 2700K but it is only a difference of 100Mhz) and at 4.7Ghz. They found that bottle necking could occur. Especially at sock but even in some titles when overclocked. Is it severe? No. Will it happen in all games? No. Will it happen with a midrange GPU like a 390X/480? Possibly.
Again it really isn't anything to worry about but making a blanket statement like that is a bit problematic. Would you say a Ryzen 7 bottlenecks GPUs? Not accusing you of bias or anything, don't misinterpret that, just curious as to your conclusions there considering at stock they generally all beat the 2600K So therefore they aren't bottleneck either which I don't think is true necessarily either..
Bottlenecking is a difficult topic to discuss or even demonstrate but it is possible across a wide range of CPUs. Even modern ones from Intel and AMD. So the idea of a 2700K even at 4.2Ghz bottlenecking a modern GPU is not at all far fetched.
He is aware.... He never was saying the 390X was faster than a 480. He said he knows his 390X is similar in performance and is thus comparing 480 results he sees to his, as many places don't bench the 390X anymore, and wondering why his scores weren't as close.
@TheCaveman summed it up nicely. I will add though that in some titles the 480 with its better tessellation performance and PDA can make it run faster than older architectures but that is limited to selective titles. That could also be a factor in the discrepancies OP is seeing and makes comparing the two GPUs difficult despite their overall similar performance.
I should clarify the statement then, as I see your point. I should have phrased it as such:
I don't think a 2700k @4.2 is going to bottleneck a midrange card at reasonable resolutions (1080p >=) in most games, as most games are going to likely hit gpu bottlenecks before the cpu becomes a limiting factor, in my opinion.
Too the point specifically about Ryzen, it is off topic so I'll keep it short. I do not think a Ryzen 7 or 5 cpu would be a bottleneck to a 390x/480 class card at reasonable resolutions in blah blah same caveats as the 2700k.
Keep in mind the GN testing is also done with a higher tier gpu than this is, idk if its a 1080 / ti or titan x, but regardless a higher skew card that is capable of a higher fps ceiling compared to a 390x/480 class card. Yes a 2700k @4.2 would bottleneck a 1080 Ti, but we're talking about large percentages slower video cards with his R9 390X.
Fair enough.
Thanks for clairfying. Didn't know if you were refering to all GPUs or just midrange ones.
Makes sense.
Edited the original post to make it more representative of that. I don't think the 2700k and a ryzen cpu are probably all too dissimilar in terms of gaming performance as the current state of Ryzen optimization sits. Obviously when games are optimized specifically for ryzen, that will change. I don't have numbers in front of me right now to substantiate, but that's just my educated guess from what I've seen/remembered. Could be a bit of a gap either way, idk for sure, its been a crazy night and I just got in a little while ago so take all statements with grain of salt lmfao.
That feel when you hit the umm.. thermal paste a bit too hard...
You are correct...
GN have done qiote a nice review of Ryzen 5, and the quadcores are slightly ahead of 2600K in gaming...
"gaming performance"
Not talking single vs multi, just fps on screen in games.
Lets not make this post into another ryzen discussion though, getting back on topic.
@OP, what kind of gpu utilization numbers are you seeing in games?
You are correct about not making this another ryzen thread. All I was saying with games like ashes of the singularity an older cpu vs newer might be a bottleneck. Though I didn't say that. AOS and even the new Doom is optimized for more cores.
Thanks for the input guys. I'll run some tools to see the kind of load im getting tomorrow .
My main problem is that on medium/high settings most games run on benches at over 60fps. Especially on 1080p (im on 1920x1200 but the difference is minor) so I was wondering seeing i seen noticeable bump up in performance when i overclocked if im just at the wall of what I can get out of my hardware.
and yeah thats why when im looking for the kind of performance i could get I look at the 480. Shouldve made it clearer that i use it as a point of reference
sorry to necro the thread a bit, turns out it wasnt my cpu that was causing issues. Amd drivers decided to use some preset to throttle to a certain fps which cause some issues and low performance. Thanks for all the help guys
Did you have radeon chill active?
EDIT: Oh wait fail, I should read that it was an older AMD card.... nvm
I think it mustve turned on by accident, when crimson came out or whatever the new panel came out I had to turn off a lot of shit i dont need