*Official* NBR Desktop Overclocker's Lounge [laptop owners welcome, too]

I will still refuse to run it on a daily basis. Not until Microcrap listens to their user base, and implements what we want! Not what they think we want, or need. If that is never, then Windows 7 will be the last OS that I run from them. I refuse to be one of the kool - aid slurping, sheep that just give in to them.

Sent from my SM-G935T using Tapatalk

 

I think I will keep the new bios dated 3/9, as I broke 24K on firestrike

 

There's one available for my motherboard too. I will try it out later and see if it's any better. I'm on A.20, new one is A.40.

 

Damn that much voltage for 5.2Ghz?

 

Experiment, but thanks for noticing the high voltage at 5.2mhz

 

I think my 1998 Motorola flip phone was clocked higher than that??

 

The 8700K that I had in my desktop before this one took way too much voltage as well. There seems to be a pretty wide variance in bin quality.

 

Cheeky John is Cheeky. Fixed

Interesting, i have 2 more coming in next week. Lets see how far they clock.

 

The board does have voltage points and what you are looking at is idle.... At least that's what the board would have one to believe...
http://hwbot.org/submission/3822005_johnksss_xtu_core_i7_8700k_2894_marks

 

So as of right now...1.375 is actually 1.47 under load on this board.

 

I hope it overclocks well! My MB with the 3/9 bios is ready for it now I hope it can hit at least 4.7ish. That would be a huge gain in performance! If it can only turbo out, then it will be disappointing.

 

I think the reason for this is Intels manufacturing process is becoming obsolete as they are pushing the envelope on it. That article that @hmscott referred us to, about how their yields for the 8700K are absolutely horrible. The video is extremely interesting to watch, and maybe he can link it here somewhere.

 

Intel Core i7-8700K Coffee Lake Z370 and Z390
http://forum.notebookreview.com/thr...ke-z370-and-z390.809268/page-41#post-10704299

IDK if that's the take away, horrible yield's, it's more that the 8700K reviewers got top tier binned CPU's (Golden Samples) for their reviews, with their OC results compared against real world (more than 1000 8700K CPU's tested) by SiliconLottery percentages of yield for each bin they sell, for not just the 8700K, but also for all their offerings.

 

That wasnt the article/video I was referring to, but ok. It stated something along the lines of only getting a low percentage of usable 8700's per wafer. That the process in which they used was coming to an end.

 

What "it"?, you don't need me to post something for you, if you have something to post that I posted, just post it

 

Lol, finding a review, or a link to a video you have posted, is like finding a needle in a haystack! I will see if I can find it. I think you linked it in one if the Ryzen threads.

Sent from my SM-G935T using Tapatalk

 

Try the 8700K thread, off-hand that's where I would have put it, unless it pre-dates that thread beginning post.

 

Found the video that @hmscot originally linked.

 

Maybe that's not the one. I will search again.

Sent from my SM-G935T using Tapatalk

 

No where near the best, but interesting all the same....
Dimm.2 M.2 Raid 0

 

AdoredTV is always going to be slanted in favor of AMD versus Intel and NVIDIA. I stopped having respect for them several years ago. At one point I thought they were good, but they always seem to have an excuse for everything instead of admitting to the failures that are obvious to everyone else.

 

@Mr. Fox Does your board have the Dimm.2 Slot?

@Trafficante

Gonna need you to redo a few of your benchmarks. Starting with Cinebench both versions.
You can not have any parts of the benchmark covered up.....

Lucky for me, I wasn't really doing any cpu benching.

 

Nope only APEX does, since it has 2 DIMM slots vs Hero/Formula/Code's 4

 

Ah, thanks @bloodhawk I was just getting ready to look it up.

I got this one for testing because it has access for the OC panel.

 

Yeah wanted the Apex as well, but it was too much hassle to run it out of the case and not being able to use 2 of my DIMM's (have 4 x 16GB)

 

I know what you mean. If it ain't one thing it's another. I basically wanted to see if this 8700K was any good and it's not. And I'm sure as hell not about to go binning them. This one came out of my P870TM1

 

Nice brother @Trafficante

 

My first was pretty decent and it died for unknown reasons. Second one was not great. Yours seems similar to the first one that died. This third one (warranty replacement for the one that died) seems better than most.

Looks like HWBOT is messed up again. It is taking hours to calculate things and a bunch of my points just mysteriously vanished.

Happy Resurrection Day everyone.

http://hwbot.org/submission/3823131_

 

Nah, actually they are getting closer to being right. You guys aren't benching in the right categories. Trophies don't count for points. (Hardware in groups where there are a lot of people do) And when they set those place holder points in place, they gave you credit, but when they slowly started fixing it those points are removed. How do I know this? Because they gave me around 900 Points for 1080N's which is 100% impossible. And when they took it away, it started showing more like right. You need to click on Global/Benchmark points tab and Hardware points tab. Before we had a bunch of 86's in there, now that they are gone your real score starts to show. (Provided they fix other parts of the system)But hay, what do I know....I'm just rambling on.

Nice score.

To do any real damage I need a 5.5 ghz or higher 8700K and I don't see me spending the time or money for that. Debated all of 2 minutes about prepping this board for LN2 and thought better of it.

 

https://www.3dmark.com/fs/15239356 | http://hwbot.org/submission/3823485_

 

This was run letting the BIOS decide what voltage to apply using "Auto" mode. Core max temp was 71°C.

wPrime 32M @ 5.5GHz - http://hwbot.org/submission/3823629_

 

Ryzen 7 2700x managed to oc up to 4.3GHz on all 8 cores right below it's max Turbo boost @4.35GHz. Amazing results Same as i7-8700K oc'd up to 4.65GHz on all 6 cores
http://forum.notebookreview.com/thr...ga-polaris-gpus.799348/page-421#post-10704818

 

Yeah leaving it on auto voltage and trying 5.2ghz is probably what blew my last motherboard back a few months ago when I thought my CPU died lol!

MSI bios overreacts too on voltage settings.

 

That doesn't sound very good. Was hoping it would turn out better than that.

Link does not work.

 

I believe he was trying to link THIS page.

 

Will probably be better. But showed results was ugly. 0.5 bin below max default boost, LOOL

Fixed link in my post... The orginal http://forum.notebookreview.com/thr...ga-polaris-gpus.799348/page-421#post-10704818

 

I do not get the comparison to a i7-8700K? CB R15 at 4.65 or 4.7 GHz only seems to show a CB R15 score of about under 1,500. Agreed only 6 cores but why the comparison?

 

I think it is more the fact that appears that it is going to suck at overclocking like its predecessor, which is a massive disappointment... again. Just looking for a reason to give AMD some love, but not finding anything to be happy about as an overclocking fanboy.

 

Why couldn't they just put all 8 cores up to 2700x default max boost? Aka 4.35GHz. Would you put 0.5 bin below max default boost for 8700K when you show your overclock? Aka 4.65GHz. It's weird.

 

Hey, water cooling enthusiasts, have you seen this prototype product, they need some backing to get it done, but it looks like it could lower temperatures to me. It's basically a water block that is designed to fit directly onto a naked delidded CPU (ie a CPU with no IHS), a delidding tool is also included as part of the package:
http://www.guru3d.com/news-story/ncore-v1-naked-the-cooling-waterblock-hits-kickstarter.html

The crux of the design in terms of it's ability to lower temperatures is centred around getting the flow of the coolant water as close to the CPU heat source (the naked CPU die) as possible - apparently it brings the cooling water 20x closer to the CPU die than current water cooling allows. The closer you bring the coolant to the CPU the more efficient the heat transfer - heat doesn't like conducting through thick layers, even if those layers are highly conductive metal, so I think this concept could work.

 

I like how you don't have to re-lid the CPU afterwards... the design makes sense since everyone serious about watercooling de-lids their CPU in the first place. That said, they haven't shown off the finned surface area, have they? With the unit itself being small, and the inlet and outlet being very close, it can't be as big as normal CPU blocks. Perhaps they've resorted to a minuscule fin width, like the cuplex kryos NEXT, which would make this cooler best used with a high-pressure pump.

Naked die cooling was already possible with MSI's Delid Die Guard, as well as EK's PreciseMount Add-on. Tests will show how big a difference relying solely on the motherboard's retention mechanism for mounting makes.

 

Have you watched the video at that guru3d link I posted - it shows him milling one of the blocks, there's a part where he mills the fins inside, might give some idea towards what you're saying about fins.

Even though naked liquid cooling was possible before, the crux of this design is that he is moving the cooling liquid 20x closer to the CPU than previously possible. It must be a thin walled water block that is small enough to be able to reach inside the socket - no need for thicker layers of metal which would slow down heat transfer.

EDIT: and a review of it at the following link too, also shows interior and fins in the video:
https://www.hardocp.com/article/2018/04/02/ncore_v1_no_frame_water_block_prototype_testing

 

I hadn't. Thanks for the tip! I count 30 microchannels, whereas the EK-Supremacy EVO has 52, and the XSPC Raystorm Neo has 54. Channel-width is of course difficult to guage from a video, but the other two I mentioned are considered non-restrictive, as you probably know.

I don't know how thick the metal of the IHS is, but the coldplate of the Raystorm Neo is actually just 3 mm. I know the distance adds up, with an IHS carrying thermal paste on both sides, but 20x closer sounds extreme. There is no arguing that this is the first waterblock to directly replace the IHS, so maybe I was wrong to think a smaller coldplate could offset the thermal performance gained by making direct contact with the die. Colour me intrigued!

 

And now we have the Jokeboard of Motherboards. This Code 92 is killing me!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
I got one bench out of it.

 

I killed the exact same board ( the VRM's to be specific) on a friends system running 4.3Ghz on the 7940X. The VRM cooling is just piss poor.

Unless ofcourse , im being retarded and yours died of some other reason.

EDIT - On second thought, i doubt yours is a VRM over heating issue.

Also here is the most retarded thing i have noticed with the newer ASUS boards (x299 Strix E and Maximus X Code/Hero/Apex) If the DP cable is leaking even a tiny amount of DC current, the board gets stuck at Code 96 and insanely picky about what DP cable you are using.
Same cable that works fine on other systems (with ASUS boards) derps out on these.

 

@bloodhawk
1: The board did not die. This janky board is still running, but will go back as soon as my other one comes in. Provided it works.
2: These are still the factory settings other than me changing the ram settings.
3: This garbage board is a temp till my Apex comes in.
4: You have to keep moving the stupid ram around to get it to boot past the code 92 error. Which stands for PCI bus initialization has started.
5: I went to Frys to grab an Apex refurbished. While standing at the podium i said I wanted to check the board. We opened the box and low and behold....A Maximus X code. WTF. How you people miss that! So i grabbed a temp board from Micro.

Not really sure how you couldn't run something like 4.3 Ghz with out the board breaking The thing comes factory at 4400....

 

Oh it ran just fine.. after about a week the system stopped booting and there was this foul burning smell from the vrm area on the top.

Thats the surprising thing, if it died running something like 48Ghz + on all cores, that would make sense. The processor was being used at 4.3Ghz on all cores for rendering and such.

When i took off the VRM heatsink, one of the mosfet's was pretty musch ready to fall off the board because of burned pads.

That exact same thing that you went through at fry's is super common for amazon purchases. People just swap out board left and right and return them. I guess the rep at Fry's who handled the return doesn't know the difference between an ATX and E-ATX board.

 

I don't have that problem. I bought real DP cables after their first set of issues many years ago. Same with HDMI cables. Them cheapy ones flicker and go black.

Also, are you talking about DP to DGPU or onboard GPU or all the above?

I have fans for the mosfets. Those can give out pretty fast if you don't watch the temps on them. Did you touch to the power for them or were they stock?

And amazon was the only place I could find one. And then I had to keep refreshing the screen till the New one came up. LOL. We shall see if they try to pull the okidoke. Also that board seems to be on clearance every where. Are they coming out with something new?

 

If I remember correctly, @Trafficante was tempted to go with the ASUS Strix Gaming instead of his current X299 Deluxe when he was buying his desktop and I suggested he stay away from it. Hopefully his new (incoming) Rampage X299 board will be sound.