Broadwell-E Overclocking results

I've been sucking up a ton of info on the web, seems these new Broadwell-E processors are terrible overclokers, especially the 6950x... I'm actually about to swap my order from the 6950x to the 6850k which not only saves me a boat load of $$$ but I would actually gain performance in games more than I would lose in my rendering solutions, at the end of the day, I will be gaming 60-70% in this rig....

I've seen the 6950x fail miserably past 4.2Ghz, the power it draws is insane probably due to the number of cores.... seems the 6850K is actually the best OC in this new family of Broadwell-E.

Any of you guys actually have experience with any of these new CPU's? Can you share some of your OC results?

Edit: Wanted to post this on the Hardware discussion, my apologies.

Edited by Xenecus - 07 Jul 2016 at 7:54am

Originally posted by DST4ME All cpus are not the same and don't worry about what others are doing. EXtreme chips are known for running hotter plus more cores means less oc speed. which is ok depending. Now having said that, the base clock for 6850 is 3.6GHz where 6950x is 3.0GHz, but both have 4.0GHz turbo which means that the 6950x is ocing more then 6850. for gaming at 4.0 -4.4GHz is a good spot to be for gaming. So think long and hard about it, but if you feel like you still want to switch go ahead.

Exactly, both will OC at minimun the same, with the 6850 having a bit more overhead and *NOT* the other way around... It's been already tested 4.5Ghz is the max that anyone has been able to OC a 6950 with the sweetspot for stability being 4.3-4.4, where as the 6850 can hit 4.6-4.8 looks like....

The bottom line is, that the 6950x "can" make sense if I was charging for my projects as it's really a professional grade chip... the best gaming cpu seems to still be the 6700 or 6800 with the 6850 giving me 40 lanes, that's about it....

I really only lose 4 cores with going with the 6850 but gain significant higher clocks (at stock and possibly OC as well).... those 4 cores are not really being utilized at all by Adobe suit... in fact after 4 cores the utilization of the rest of the cores is negligible at best.... I actually think the best bang for the buck is the 6800k on these Broadwell-E CPU's unless you're going 3 or 4 way SLI then the 6850 is a must.

Not sure there's anything else i'm losing by not going with the 6950x, since Turbo Boost 3.0 is a feature in all these new CPU's

Edited by Xenecus - 07 Jul 2016 at 9:45am

Originally posted by DST4ME Look at how at 3.0GHz 6950 keeps up with the 6850 at 3.6GHz on productivityon most stuff and that is with significant less clock speed .6GHz.

That is the bottom line... the point being in reverse, look at how the 6850K keeps up with the 6950x at $1100 cheaper... Makes sense?


And I completely agree about the OC... I like to OC as higher as I can go fro bragging rights and just that, I will probably keep my OC @ 4.4Ghz Max.

Originally posted by DST4ME You missed the point, the x chips are for enthusiast or powerusers, if with .6GHz its almost keeping up, then both at 4.0GHz the 6950 kicks 6850's butt in productivity, and yes there is a price you pay for that, whether the price is worth it or not is dependent on how important the work one is trying to do is, aka is doing this work faster worth the extra $1100 to me? If the answer is yes then one goes with 6950 is the answer is no then 6850. Again please stop thinking you know your oc from DS, you are going to be disappointed, stage 1 os is upto 4.4GHz they don't promise nothing, depending on your chip you may not even get that far no matter which you pic. Plus there is other factors, I rather keep a oc of 4.0 with vcore below 1.4 then I would an oc of 4.4 with a vcore of 1.48. DS does not pay too much attention to vcore.

I'm not counting on DS OC to it's full potential... I never seen a boutique master the art of OC, I know what they do is just stage settings from the BIOS for the most part... i totally get it. I'm going by enthusiast that know their OC and that actually tested these chips, that's all. I respect your opinion, make no mistake, but the X or the K mean very little aside for just that... at the end of the day they are both unlocked chips, right?

Also keep in mind in the real world when running at all cores the 6950x WILL NOT run past 3.5Ghz in turbo mode that alone 4Ghhz it will ONLY reach 4Ghz in single thread... in multithread test it rarely goes over stock speed...so lets get that out of the way.

Edited by Xenecus - 07 Jul 2016 at 10:41am

Well I'm discussing the best overall CPU.... and I'm finding it hard to see how the 6950x is better than 6850k... Besides the obvious reasons of the 4 more cores, then again is 4 more cores worth 1100... that's where I say no....

Having said that, why would the 6950X @ 4.0 have better performance that the 6850X @ 4.0 generally speaking? (real world not synthetics)

Can't argue with your logic :)

I much prefer Broadwell-E though, Turbo Boost 3.0 and 40 PCI-E Lanes are the features I was after (Assuming at some point I will try 3 way-sli)

Don't buy into Turbo Boost 3.0. And like DST4ME pointed out, tri-SLI configurations are no longer supported (except for select benchmarking apps, with no real world application), so 40 PCI-e lanes really aren't needed.

http://semiaccurate.com/2016/06/13/intels-broadwell-e-not-released/

Originally posted by SemiAccurate Luckily for the Intel marketeers, there is a solution to this clock speed mess, having a vastly cheaper part that is better in every way then your flagship simply won’t do. To rectify this, Intel invented, “Intel Turbo Boost Max Technology 3.0”, something that is called out separately from Turbo 2.0. Why? Because it isn’t real, it is a complete marketing sham that does next to nothing for the user. Intel promised the moon for the new Turbo Max 3.0 and then didn’t go into how it works. SemiAccurate asked about how it worked and was given a useless quote from a technical guide Intel sent to others but not SemiAccurate. We asked, “The new turbo, when it ID’s the fastest core, how does it assign critical processes to that core? Software? Firm/hardware? Is it Windows only or OS agnostic?” The reply was: “Intel Turbo Boost Max Technology 3.0 works by using a driver coupled with information stored in the CPU to identify and direct workloads to the fastest core on the die first, allowing workloads to run more quickly and efficiently. The driver also allows for custom configuration via a whitelist that gives end users the ability to set priority to preferred applications. The driver MUST be present on the system and configured correctly in order for this benefit to be realized as current operating systems cannot effectively route workloads to ordered cores. People can then use Intel Turbo Boost Max Technology 3.0 Application to further optimize. I’ve attached an addendum that further explains the options once the application is running.”(Note: Intel didn’t actually attach anything to the email.) Since Intel declined to answer the questions we asked, SemiAccurate dug in. Max 3.0 is a driver that only works with Windows, it isn’t actually a hardware or board feature like Intel wants you to believe. It only works with specific boards, specific firmware builds, and with specific drivers, none of which are available as a single package. Good luck to the average user trying to get it all to work, and don’t expect it to be supported once the launch party is over. Should it work you are given a text box which you can input a list of programs that get priority. Intel says the program, sorry, feature in marketspeak, will identify the highest clocking core and effectively pin the hardest working thread to it for maximum performance and boost. How Intel identifies which thread should be pinned would be interesting to hear about but Intel refuses to actually talk tech so we can’t explain it. As far as we can tell you can only list the process which may or may not have the intended results. In any case if you don’t the ‘technology’ will pin the task in focus automatically. Intel claims that it is, “more than 15% faster” based on SPECInt_base2006 running on a 6950X vs a 5960X. We feel obliged to point out two things here clocks and marketing. Since SPECInt_base2006 is a single threaded test and both listed cores clock to 3.5GHz, that number should reflect the IPC differences between the Haswell and Broadwell cores. That difference is maybe 5%, 15% is right out. If you read the fine print on the slide, you will see that Intel recompiled the binaries so that 15% difference that they are attributing to Max 3.0 is likely 2/3 due to compilers, 1/3 due to core microarchitecture, and 0/3 due to Turbo Max 3.0.

Edited by  - 07 Jul 2016 at 5:56pm

Correction:

1. nvidia does not officially support tri sli so that is a problem, you can get the "enthusiast code" but that supposedly will just unlock the tri sli ability, it does not mean the same thing as nvidia supporting tri sli.

2. there is no hb bridge made for tri sli as of right now, you can go ahead adn use the old one but they are not hb. If HB is not needed then why would nvidia say it is for dual sli.

3. lets say regualr sli bridge didn't make much of a difference and one did get the "enthusiast code" from nvidia, do we have a code that tells the game to use tri sli?

Don't bank on things you see and read on the net so fast.

Lets see if nvidia does in fact provide a code that works as it is supposedly suppose to, and everyone can use this code, and then lets see what happens with the bridges and if those 2 workout lets see what happens to the games themselves not to mention dx12 and ms.

At the end of the day, if nvidia is not supporting it, it means that game devs will not waste their time on it either.

Edited by DST4ME - 07 Jul 2016 at 7:59pm

Honestly I didn't keep up with it as it didn't make much sense to begin with.

Can you show me a tri sli with regular bridge vs hb bridge benchmark? both in 3d and actual game.

I don't know one informed poweruser/gamer that cares about what a Synthetic benchmark is on a part, everyone I know cares about real world performance. How many times have you seen us mention benchmark numbers when we recommend parts? Zero and there is a reason for that. Application benchmarks are what counts.

IMHO with all due respect for an average/ordinary owner user, Synthetic benchmarks are worthless.

Only people in the industry that have something to gain from breaking benchmarks or having high benchmarks, no informed person is gonna recommend gpus based on 3dmark bench score for example.

I'm not saying there is something wrong with benching your system, but to spend $500+ on things that have no real world use just to get a higher bench score is not gonna grab anybody's attention or respect.


I guess what my long rant up there is trying to say is this:

If its about benchmarks for somebody, then that is great and fine, but as long as that person understands that its about benchmarks for that person and a small group of people, not the rest of us.


So to each his own.

Edited by DST4ME - 08 Jul 2016 at 9:11am

http://www.pcgamer.com/the-broadwell-e-review/

Jarred is a former Anandtech writer and one of the most knowledgeable with processors. He makes a lot of points and it his conclusion, it's the same as what we've been recommending on this forum: pure gaming you go with an i7-6700K or i5-6600K; pure video editing or other professional multi-threaded application you go with a Xeon workstation; and a mix of gaming and professional multi-threaded application you go with the HEDT processor. Gaming performance is pretty much a wash, as far as Intel processors are concerned, and multi-threaded applications get cleaned up by the processors with more cores.

Edited by  - 08 Jul 2016 at 9:54am

The CPU RND has slowed down a ton in the past 3-4 years compared to GPU.... There used to be a staggering delta in a 3 year timeframe just a few years back, seems we have hit a wall in speed/cores, I'm guessing that consumers mark the trend and four cores seems to be where software companies seem to focus on.

I agree having 3-4 way SLI is a complete waste however there are a ton of folks out there that have these expensive hobbies of ranking their benchmarks... who am I to judge? :) Just as there are people that would buy a 6950X (hell I almost did, they I realized that for what I use a 6850 or even a 6800 would be just as good if not better due to the higher base clock.

i agree, to each its own, its a free world :)

It's not about bragging.... i don't think any of those guys care about gaming they do it because they can for one and they like benchmarking... I mean you would buy a 1700 dollar processor which makes very little sense to almost anyone, but you are entitled to do so if you can.... having a 3 way SLI is just as unnecessary as having a 6950x... that's all am saying.... That's just what I think, not saying it's what you should think.

Some people buy a Ferrari to go from point A to point B, some are happy with a Civic ;)

At the end of the day you and I both bought a DS system, I have a long list of people that tell me again and again how they can do it for $$$ less... Hell I can build the system I bought for a lot less, but to me it's worth it for my reasons...

Edited by Xenecus - 08 Jul 2016 at 3:22pm