Why no love for the Ryzen line in build of the month?

[Lathais]

Just to quote you;

'In short any decent desktop quad-core I5 can run pretty much everything you want and game at the same time with little issue, no need to make up shit really'

Fun fact, the video I posted isn't even making shit up.

Well, it kinda is making things up because who would be watching a video at 8k on their 4k monitor while gaming. Unless they have dual screen 8k displays, it's a pretty made up situation.

[Deleted]

Well, it kinda is making things up because who would be watching a video at 8k on their 4k monitor while gaming. Unless they have dual screen 8k displays, it's a pretty made up situation.

In fairness thats not quite for you or me or the poster I quoted to decide really, if you have the hardware grunt, well theres a reason to use it otherwise it goes to waste, its actually a good way of watching high quality content, even 4K 60 FPS footage is CPU heavy and you can still leave room for the GPU to run games if you want.

I proved him wrong and provided there is content out there that shows what there here and now can do to current hardware, its also a good way of showing what future proofing looks like.

[Life-Binder]

Ryzen already gets too much "love" from its frothing fanboys that think its the fastest CPU ever made

The single core performance of the 7600k barely beats out the 1600x

no, it beats it by quite a lot .. especially OC vs OC



the MMO-C builds are pretty much only for WoW .. for WoW they do list the best CPUs which are always Intel

[Holofernes]

lol that vid. put to 8 k resolution its consuming 30-40 % on my ryzen 1700 system . wtf is this ? want to see a intel 7.7 k guy report its utilization

[Life-Binder]

and yeah all the typical background stuff does not put a dent in any modern quad


the only activity that may eat into a CPU is something like decoding/video render related while gaming or streaming

[Lathais]

Ryzen already gets too much "love" from its frothing fanboys that think its the fastest CPU ever made



no, it beats it by quite a lot .. especially OC vs OC



the MMO-C builds are pretty much only for WoW .. for WoW they do list the best CPUs which are always Intel

I don't think even the fanboys are saying it's the fastest ever, but it is true that it only beats it by a little, not a lot, even OC for OC.

http://www.gamersnexus.net/hwreviews...xciting/page-2

There you can see OC vs OC. It's only really behind in certain synthetics.

Take it to gaming though:
http://www.gamersnexus.net/hwreviews...xciting/page-4

And that 1700x with an OC has better min frames than the 7600k with an OC in Warhammer.

In BF1 it's only about 10 FPS behind, though intel does do better on min FPS here.

You are a smart guy, you can look at the rest yourself and see that it is not drastically behind. Yes, it is a little behind, but not as much as people make it out to be.

What we "fanboys" do say though, is that a 1600 is much cheaper than a 7700k and it can get pretty darn close in performance. It may not be the fastest CPU ever made, but it has a phenomenal price/performance ratio.

[Deleted]

Another video to test for people, for some reason this one pegs the cpu a bit more;

[Tiberria]

I mean it's not crazy to think that people will watch 1080p and eventually 4k videos on another monitor while gaming, especially in a game like WoW that tends to have a lot of downtime built into it.

Just as a test, I turned on a 4K YouTube video in a seperate Chrome window on my second monitor with WoW running. With this video, and 7 other Chrome tabs open, I am at 17% CPU utilization from Chrome - on an R7 1700X @3.8 GHz. That suggests that Chrome alone is eating up 1.4 of 8 CPU cores. If the data in that article shows that WoW performance drops when you have less than 4 cores available, wouldn't using 1.4 of the cores on Chrome and a video you're watching on another monitor indicate that a quad core CPU (or at least a quad core with just 4 threads) would see performance degradation in WoW from this? Yes, I realize that 4k videos/streams are rare right now, but are going to get more and more common as monitor resolution and bandwidth gets better.

[Gray_Matter]

Clearly represents a realistic background workload when gaming.

70% with 4770 non-K @3.7

What is realistic? My daughter normally has a movie or Netflix playing in the background, along with a web browser and a paused minecraft all running in the background while she plays a game. I have 2 instances of Firefox (32 bit and 64 bit) along with a virtual machine, Outlook and sometimes an encode.

The only thing that isn't realistic is what the guys doing the benchmarks do which is to have no background tasks running at all.

[T1berius]

if single thread performance is the most important thing then yes, an i5 7600 is better for you buck than a ryzen 5 1600, however, if single thread performance is only tertiary and in fact multicore functionality has the capability for making up for single thread performance, the ryzen 5 crushes the i5.


so please for the love of fuck can someone show me evidence that wows performance is tied directly to single thread performance above all else AND that multicore support is "meaningless", and we can finally put this to rest


because it's been more than a week and all I've seen is the same german website posted over and over as the ultimate evidence


my personal ryzen 1700 single thread score is about 15% lower than an i5 without an overclock, but it's overall benchmark is almost 40% better, so lets stop with the "haha u don't know shit intel rulez amd droolz", be mature about this, and get some real evidence in here

[Kagthul]

lol that vid. put to 8 k resolution its consuming 30-40 % on my ryzen 1700 system . wtf is this ? want to see a intel 7.7 k guy report its utilization

Substantially less, thanks to the native decoding via the iGPU

[Lathais]

if single thread performance is the most important thing then yes, an i5 7600 is better for you buck than a ryzen 5 1600, however, if single thread performance is only tertiary and in fact multicore functionality has the capability for making up for single thread performance, the ryzen 5 crushes the i5.


so please for the love of fuck can someone show me evidence that wows performance is tied directly to single thread performance above all else AND that multicore support is "meaningless", and we can finally put this to rest


because it's been more than a week and all I've seen is the same german website posted over and over as the ultimate evidence


my personal ryzen 1700 single thread score is about 15% lower than an i5 without an overclock, but it's overall benchmark is almost 40% better, so lets stop with the "haha u don't know shit intel rulez amd droolz", be mature about this, and get some real evidence in here

Will it utilize more cores/threads if they are there? Yes, no doubt about that. It just tells Windows I need this thread run and it assigns it a core and it goes on to the next. The interesting part becomes in that if you take something like an i7 and disable HT and all but 2 cores, performance just doesn't change. If you then take those 2 cores an OC them, performance goes up. From this, it is evident that 2 strong cores is more important than more cores to WoW. One core to handle the massive draw call thread and the other to handle everything else. Just because Windows divides the work up among all your cores does not really mean that all need to be being utilized.

[Linkedblade]

Substantially less, thanks to the native decoding via the iGPU

No iGPU on kaby lake-x or skylake-x. Stupid change really.

[Kagthul]

No iGPU on kaby lake-x or skylake-x. Stupid change really.

Relevant to the discussion... how?

We were talking about Coffee Lake performance. Which has an iGPU.

[Linkedblade]

Relevant to the discussion... how?

We were talking about Coffee Lake performance. Which has an iGPU.

Uh, well coffee lake-x may not have it either if they keep with this new standard they are putting out.

[Kagthul]

Uh, well coffee lake-x may not have it either if they keep with this new standard they are putting out.

Again... irrelevant. The discussion was about comparing non-HEDT parts.

[T1berius]

because i like evidence so much, here's some more from me

this is my highest stable oc with the ryzen 1700, at 3.85ghz, compared to equal intel systems

http://imgur.com/a/oSkDv

single thread performance, the ryzen 7 lags significantly behind (with an unstable 3.9ghz oc I can get that score up to 1905, but I don't think it can be run stable that high)


http://imgur.com/a/9mv1v


and full performance, ryzen pretty much outperforms all around

[Sorshen]

I think you're being intentionally obtuse at this point...

[T1berius]

I think you're being intentionally obtuse at this point...

well i just got another 7% bump in single thread performance on a lower oc on my ryzen 7 so idk, maybe i'll find some settings that can get me another 20% which is all I need to completely discard all of this because it will be moot with a better single thread performance

[mrgreenthump]

I think you're being intentionally obtuse at this point...

He'll fit right in then.


But, really.. This discussion is getting pretty silly and somewhat off-topic. Team blue and green both using numbers/tests that suit them, ignoring everything that might shatter their illusions. People using somewhat unrealistic usage cases for CPUs.. Who really is going to run any type of a youtube video while he is raiding in WoW and being serious about it?