Jump to content

Yes, I know. I know. I promise you, I know.

 

What I don't know is: by how much?

 

Here's the story: I'm itchy for a rebuild. I watercooled my current rig to scratch that itch in January (and have since upgraded the loop, even). But now, Polaris and Pascal are out. Zen is around the corner. Keeping up with the TechTubers has reminded me that cooling just isn't cool enough. My rig is still old and I still want something new. More importantly, I feel like the RX 480 hit the nail on the head for price to performance as well as addressing an under-fed market segment... but it just isn't fast. And who knows when Vega will come out or whether or not it'll be properly competitive. October? Then what about Volta? Kaby Lake? Of course, this is how it goes. I've been an AMD supporter since the K6-2 and was hoping to do my whole next rebuild with their new, exciting stuff, but Polaris 10 just doesn't seem to cut the mustard when I'm coming from two overclocked 770's. Unless I went with two 480's, but the current benchmarks/apparently dismal CrossFire support is dissuading me from that course of action.

 

And then I read this article last night. And it really threw me for a loop. But before you say it, this is what I was on about in the first line of this post: I know Xeons are server chips and games prefer clock speed and la-dee-dah. I'll make it worse for you: I don't even do video editing, F@H, or anything to justify such an incredible amount of parallelized computing. And yet here I am, considering building it anyway, fully intending to play games and browse the web on a dual socket server board. Oh, and I'll totally install the multi-core variants of gzip and lzma on my Arch drive just for giggles.

 

But here's the point of this post. Given identical GPU's (say, a 1080 FE. Or a 480, in my case) what would be the impact on framerate averages, minimums, and timing between a pair of E5 2630 v4's at 2.2GHz and a 6700k running, say, 4.5GHz? Or what about disabling 7 or 8 cores on each Xeon in an effort to keep the temperatures low and ensure the 3.1GHz Turbo clock would be utilized, still having 4-6 cores available on the machine, and comparing that foolishness to an overclocked 6700k?

 

tl;dr

 

I know a 4.5GHz 6700k would produce more (quantity, consistent, etc) FPS than a 2.2GHz E5 2630 v4. Of course it will. But I want to know if anyone else knows: how many more?

 

I'll also happily accept an educated guess, more clock-for-clock comparable models (4790k?) etc. Just give me something to chew on, internet. =)

Link to comment
https://linustechtips.com/topic/640904-xeon-e5-2630-v4-versus/
Share on other sites

Link to post
Share on other sites

They used ES cpus which are stolen, may fail, and may not work on some boards.

 

The framerate differences would be huge between the xeon cpu and an i7 because games simply just fon't use the threads, I have a 16 thread cpu and have hit 65% usage max in gaming. You are throwing money away by even considering adding a second cpu for gaming.

 

Disabling cores on the xeons wouldn't help usually when 4 threads are used the cpu will be only 2-300MHz higher in turbo boost speed, the 3.1GHz is only for single core usage.

 

A newer gpu like the 1080 will suffer from having the xeon, not due to threads, but from a lack of clockspeed on the cpu.

 

if you want to spend $400 on cpus get an i7 6700k and oc it, you will have a much better experience, or spend that money toward a 1080. 

 

 •E5-2670 @2.7GHz • Intel DX79SI • EVGA 970 SSC• GSkill Sniper 8Gb ddr3 • Corsair Spec 02 • Corsair RM750 • HyperX 120Gb SSD • Hitachi 2Tb HDD •

Link to comment
https://linustechtips.com/topic/640904-xeon-e5-2630-v4-versus/#findComment-8261636
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×