Intel Core i9-9900k 8c/16t, i7-9700K 8c/8t, i7-9600k 6c/6t 2nd Gen Coffee Lake CPU's + Z390

Those are both more inline with what I saw, which is around 200W at 5GHz. First one shows 205W in blender (with HT, Without HT it did 5.2GHz@195W), while he saw the 8700K hit [email protected]. (kit guru isn't my go to on benches, and I think got called out initially on benches for something on overclocking a year or two ago, but likely has been addressed, so those numbers are fine).

The other one I'm not familiar with (Only know certain European overclockers, so not doubting his results) got 209 on CPU package at [email protected].

But both of those agree with what I've said and the ranges I've given, so I don't know what you are trying to do responding other than giving more data that falls right inline with my ranges and power draws. If just trying to give more data to the community, then thank you. If trying to stand for a different proposition, like the comment from the reviewer at Kit Guru on power draw, then I have to ask for the data from power draw on the laptops using the 8700K at different overclock frequencies. I'd do it myself, but after Clevo decided not to let ZM owners upgrade the GPU and outdating our machines within 6 months, I just went to a desktop and will likely not buy a DTR for my next laptop (which the reason I got one was to not have two computers anymore, they did similar with my HM model).

 

Stock 9900K and stock voltage. Blender with TDP tuning (you should be able to shave of +40w with undervolt) Techpowerup.com

 

That's cool, [email protected], 180W at 4.8GHz all core, on a Maximus XI Extreme board. Good stuff and we are starting to see that pattern on volts and wattages. This is another review I did not see, but have seen W1zzard's work before on OCing, so should be solid data. Good stuff!

 

The MSI GT73VR trashbook shuts down hard if you bypass the current limit.
I don't know how many amps it is.

Don't know what it is, but seems to be a hard coded trip.
AVX Prime95 at 4.7 ghz @ 1.265v small FFT triggers this in 1/10th of a second. Can't even see the power draw it happens so fast.
With AVX disabled it can pull 103W with no problem (but gets hot=90C).

Setting "PS Current Threshold1" to 300 (maybe higher than 255, i don't know. 255 doesn't cause it. max limit is 512, which corresponds to 128 amps as its a 1/4 divider) causes **NON AVX** loads to shut off the laptop, but it takes about 1-2 seconds to happen.

 

FYI - i9 in P870DMG and F5 / MSI 16L13:

Offical Coffee Lake Thread/Upgrade Guide MSI 16L13 (Eurocom Tornado F5)
http://forum.notebookreview.com/thr...urocom-tornado-f5.821651/page-8#post-10812686

Interesting followup posts...

"Its definately a bios issue and honestly, I'm not to bothered as I was going to run a 8700k in my clevo anyway. There really is no need for a 9900k in a laptop when it can't be adequately cooled.

My friend has a 360mm nzxt rad with METAL PASTE on the 9900k at 4.8ghz and on prime95 it reaches 92c in about 10 seconds.

So, if a 360mm rad can't cool an i9 what the hell is going to cool it."

 

i wouldnt want to get x299, but the x599 one, because if i go big then i do as big as possible, otherwise i'd wait for zen2 for a much better performance/value for like a 5-10% cpu performance difference in single threaded scenario.

also, getting a 9700k vs 9900k will be dependent on the bin and cost. 9700k is actually priced about 8700k so a good bin with low price is probably much more worth it than 9900k binned chip. ultimately deciding factor for everyone would come down to price what they're willing to afford. if im going to disable HT on 9900k just for the extra 4 MB of cache, then pricing reflecting that better be justifiable (in my own standard).

im guessing a 9900k 5.2 or 5.3 binn would cost several hundred dollar more than 9700k binned at the same frequency, which is absurd for me this time around because HT will basically render my machine unusable at high clocks due to heat.

we know intel will tell us, "use cpu at 1.4v to 1.5v, its good for you!"

 

It might be possible, but using Prema as a bellwether, P[7|8]*DMx shouldn't be too much of a surprise. P[7|8]xxxTMx seems to be a different story - http://forum.notebookreview.com/posts/10812711/

 

"H" Processor line 4 core GT2. (45W), 86 amps, 86 * 1.25v = 107.5 W.
I don't know how this relates to the BGA turdbook shutting off.
I can draw 115.2W at 4.9 ghz cinebench but the laptop just shuts off after about 92C and instantly power cycles itself (shuts off and turns on again).
I don't know.

1.38v * 86 amps = 118.68 watts.

who knows.

 

That 28-core will be a beast, but the price tag, if rumors are right, will be likely $3500 for it (and need LOTS of cooling). At the 9th gen Intel event, PCWorld reviewer lined up the Youtube reviewers asking them to guess the price, then revealed that the Intel reps he talked to said it would be much less than $5K. Considering the core count being 57% over the 7980XE/9980XE, I could see $3K to $3.5K, but wonder the cost on that MB to go with it ($500-1200). Bad value, but amazing engineering!

For the Zen 2, I estimate a 24% performance over current gen, based on a guess of 4.5GHz all core boost (about 10% on speed) and 13% IPC (the two rumors are the 10-15% additional IPC depending on workload and the 13% in math and science workloads leaked recently which was seen on the server chips with no data on speed given). If that is the case, then on mainstream you would get a couple percent over the 9900K, on HEDT, that is in the 4200-4500 range on CB for the 16-core chip, which would beat the 16-core Intel offering, and should just be incredible for a 64-core chip on server side. I analyzed this the other morning, but there is a caveat on HEDT that doesn't apply to the server chips estimate or the mainstream.

So, we know the coffee lake chips are 14nm++ for mainstream, and we know the Cascade Lake-X/SP chips will be on 14nm++. But, the 9th gen HEDT chips are said to be Skylake-X again. Reporters are saying it is 14nm++ also, but Intel has been changing the name with the process revisions of 14nm and the Skylake-X was originally on the 14nm+ line (kaby lake process). That means that releasing the HEDT version on 14nm++, if AMD just gets 24% additional performance, it could keep the crown, even if the value proposition isn't there. That is a lot of speculating on my part, but it is within reason.

 

on hedt, intel basically lost their advantage unless exotic cooling is involved. no frequency gain then its all about core count, and AMD has the value proposition best for consumer at high end, cost about half of intel, rest can be spent on a good cooling solution while getting similar or even better performance due to the cooling solution.

right now if i were to go for AMD HEDT, i'd wait for zen 2 or possibly zen 3. 2990wx is a bit of a disapointment because unlike eypc, 2990wx die missing direct pcie/imc connection which is a kill joy. zen 2 looking really good if rumour turns out to be true, then zen 3 will simply replace it with same socket.

some of my biggest concern with AMD is that no TS support, and the new overclocking software forces people to upgrade to windows 10. the storage performance will be less of a concern, because i'll be using optane SSD at that point and most of the application can't really take advantage of it. i can list a few, overall its okay to sacrifise storage for CPU performance.

 

There were some really cool videos on optane on AMD systems decently recently. I know Level1Techs did something with it and think LTT may have.

In any case, I do agree, on HEDT, Intel lost the value proposition at TR1 and TR2 was just insult to injury. Even with finding ways to mitigate the memory on PCIe direct connection issues, it would have been nice if it was connected. That helped Intel keep the pricing they had where they had it. But, in less memory intensive tasks where you can hold it in cache, like rendering and certain other tasks, those 32-cores are just phenomenal! Intel's 28-core Xeon, though, won't have the memory issues and is getting hexa-channel memory, making me hope AMD counters with octo-channel memory on HEDT and 16-channel memory on the server side (talk about mind blowing)! I really hope they do a memory bandwidth race SO BADLY!!!! But time will tell.

And, yeah, I asked Webb about throttle stop support, he plainly and directly said no. OC software for AMD currently SUCKS!!! So bad. Good UEFI is needed because of that. Really wish I could try out Precision Boost Overclock, or doing a fusion like the EVGA Dark auto OC but with X399 or X499 boards. AMD's CPU step downs are similar to Nvidia's auto OC, but more granulated, so if they could add AVX offsets and allow for the BIOS to auto-OC while differentiating on instruction sets, talk about a one stop shop for auto clocking based on your cooling solution for consumers. Now that would be value added times 10! Something I thought about recently but is likely a fart of a prayer.

But, yeah, that is why I would wait on HEDT to confirm all they did was add solder if on Intel (at which point I'd recommend not going from 7 series to 9 series), unless it is a non-OC machine for production, then it might make sense to get that solder. Instead, wait for next year for like Cooper chips to drop. (edit: or consider Zen 2)

On mainstream, I see purpose uses for 8700K and 9700K, think 9900K is bad value, and see purposes for the 2700X.

Unlike anytime in the past decade, between both AMD and Intel, there is something for everyone, at all price points and purposes. I am REALLY loving that!

 

Primo Cache can perform the same functions as Optane and works with all SSDs on any platform.

 

wait didnt AMD already have octa channel? or is that only for Eypc? maybe i remembered wrong i thought AMD TR are similar to Eypc but i guess they removed the extra channels and direction pcie/imc connection. gotta pay more if we want more.

TS is really a let down, its really easy to use but no such thing for AMD. another reason to wait it out for zen2 or 3 is possible AVX2/512 being included. also i'd be interested if AMD have something similar to intel's hedt adjust individual core voltage (per MSI) also turbo boost 3.0, like have a software that runs off 1 or 2 cores that actually work.

btw, the game mode, can it be switched without restarting machine?

 

i had a long discussion with mega whats his name, primo cache sucks. it doesnt give user the control in comparison to a ram disk, and even with lower access time, your computer chooses what to cache. though i do use primo ramdisk software, its great.

also, optane SSD has its use, its just that as faster the technology is, benefit becomes less transparent.

 

So, Epyc is 8 channel, TR is 4 channel, but winds up with better memory bandwidth than Intel unless you have very high clocked ram with decent timings and high core clocks, then it switches to Intel. For game mode, I'm on gen 1 and yes, it does need restarted for that, but no, it isn't worth restarting the machine for the gain. Just set it up and run as is. You can run channel interleaving if you like for higher peak throughput, but generally I get better performance just letting it do its thing. Windows handles NUMA decently well for CPUs. So I don't bother with messing with local vs distributed, just leave it with seeing it as UMA.

And TS not working with AMD is disappointing. Ryzen Master is getting better, but I still prefer doing it in UEFI. Call me old school. When I find a program I like using, then I will. But it is like hating XTU, DIP5, etc. (although DIP5 is better than Asrock's on AMD by miles).

As for AVX2 and 512, it depends on if you use it. Professionals do, most end users don't. When you need it, you need it though. So I'm hoping they do something for those and an offset.

On the core voltage, HWInfo can read the per core voltage, so that sort of monitoring was put there for a reason. I was hoping to use that and set voltage levels not just on per core to keep the chip as cool as possible, but also to set it at different boosts. AMD added the star to show which core in each CCX is fastest and second fastest, and identifies the fastest core out of all CCXs (or best core). I think the logical next step is doing that to distribute the heat while boosting the best cores in the CCXs at lower voltages, which may be how the Precision Boost Overdrive works, which even Der8auer said was good enough that it should be used on CPUs with it when he was overclocking the dual socket rig for gamescom. So, soon, how Nvidia mentioned the one click OC, we are likely to see further refinements on AMDs side for auto OC which will take away the fun manual OCers have (unless those features are given to OCers in the future, rather than just P-state OCs). But, one step at a time.

But, Zen 2 could just bring those advancements, or do an auto-OC where you run it for 20 minutes to 4 hours one time once setup and be done with it (and let the person control how long to test or the level of testing or something like that). Just got to wait and see. Also I like the thought of allowing manual clocking of all states and voltages for stability as I'd have fun getting it dialed in to a T. But I'm an odd duck like that.

 

Probably the chips being unstable. Overclocked at the border and chips pass +86C (the temp limit der8aour say you risk unstability with high overclock).

 

Chip isn't unstable.
The laptop just power trips. It's not a BSOD or even WHEA correctable error. it's a "trip".

Aka turns off and then turns on again in half a second. But it only reaches this past 92C at 4.9 ghz. At lower mhz, lower volts, can reach 100C (without AVX). Remember that indian guy spoke to the MSI engineer and said there are some hardwired amps/temp cutoff points depending on things he didn't understand (although they were talking more about the GT83VR trashbook).

(exact same thing as trying to run AVX small FFT prime95 @ 1.265v @ 4.7 ghz, except this happens instantly).

 

Put it this way... Try increase volt for 4.8GHz so it pass 92C equal as your (4.9GHz overclock run - shuts off after about 92C). Will it then pass? Do the same for 4.7GHz.

 

86A is 8th gen. For 7th gen H-Processor Line (45W) - Quad Core GT2, default core/IA ICCmax is 68A: https://www.intel.com/content/dam/w...s-datasheet-vol-1.pdf#page=105&zoom=100,0,220

Keep in mind that package power is a total number including IA cores, GT cores (iGPU), system agent/uncore and on-die DRAM controller, so the power going to just the CPU cores is a bit less than the total. Nonetheless, your core ICC is well over 68A, especially at >1.35V, since current is directly proportional to voltage (Ohm's law). IIRC you set ICCmax to 200A?

Since your laptop shuts off after 92C, it could be VRM OTP (overtemperature protection) kicking in.

On Clevo -TM boards, the current limit throttling happens immediately after starting a stress test that exceeds 138A, even when temps are low.

 

I'll try that later.
But the MSI engineer told beastforever, the dragon, that there are "Zener doides"that are set to power trip the laptop under certain conditions, depending on power draw, and that two trip points were 92C and 94C.

It's not the VRM's tripping because of temperature either. The VRM's are frosty. (literally an ice cold room).

The exact same trip happens if I run 4.7 ghz avx ENABLED small FFT prime95 @ 1.265v., except it happens INSTANTLY.
Remember I told you guys you can force this exact "power trip" instantly?
(Doesn't work on MSI F5).

Download RW EVerything, go to EC RAM Register D7.
Put an 01 into the register.
The laptop will instantly turn off and power on less than half a second after powering off.

Put a value HIGHER than 01 into this register (02 through 7F), it will count down like a BGA bomb, byte by byte, until it reaches "01" then it shuts off and turns on again.
(80 seems to be the same as 00, 81 seems to be the same as 01, note 80 hex=128 decimal)

 

Hummm, that was all an interesting read to say the least, but I'm still getting a 9900K and will be running it in this P870TM1.

 

Please post video of the "smoke test".

 

XOCer sees heat and power constraints on a platform and says "challenge accepted." Looking forward to it. Let us know of any mods you wind up with, etc.

 

I hope they keep making them this way. That way they stay out of the hands of these so called lucky overclockers. Where all the chips do the same clock speed whether they are on air or LN2. (Sandy Bridge) Then came Ivy Bridge and they all went back to be less than average overclockers.

Totally!

 

Spoiler: It's Good to be King

 

Man, I wish, but i'm only ranked 20th in the United States. So in that respect....I'm still a peasant.

 

Still waiting on my chip to come in....

 

I should have mine in the morning/early afternoon. 9900K that is.

 

ill be sure to follow in your footsteps and see how far i can push the 9900K in my trusty old DM my goal: achieve stock clocks without my machine going up in smoke

 

Nice.

I hear ya. This one was quite toasty the very first boot, but after that it kind of got in some sort of line of order.

 

i9 9900K Overclocking, Temps & Silicon Lottery!
Joker Productions
Published on Oct 25, 2018
Overclocking the Intel i9 9900K on my golden sample that can do 5GHz at 1.25v and even 5.2GHz or 5.4GHz.. however the temps can get pretty toasty. Do you really NEED a water cooler for the 9900K?

Spoiler: Click to See Video

The i9 9900k Gets Overclocked & Wet!!!
eTeknix
Published on Oct 24, 2018
With a lot of people commenting on how hot the i9 9900k can get, we wanted to put it under a custom loop to see how far we could push it and how high the temperatures would really get. By getting the i9 9900k wet, did it still run hotter than hell and how far were we able to push it? Watch till the end to find out!

Spoiler: Click to See Video

i9 9900K Thermals & Voltage Featuring The BeQuiet Dark Rock 4
Tech Showdown
Published on Oct 24, 2018
So many people are talking about the high Intel i9-9900K Temps and how each reviewer was testing the CPU thermals, today I lay it out.

Spoiler: Click to See Video

Intel 9900K Miami Launch + RTX 2080Ti preview (Vloggish)
Level1Techs
Published on Oct 25, 2018

Spoiler: Click to See Video

Good news for IOMMU users
0:00 General Update, Sup with RTX2080ti?
2:54 Z390 coverage

 

Still haven't reviewed one of the videos, but nice seeing more confirming rough voltage and wattage.

I wanted to point something out on eTeknix video which people may or may not realize, when you cool your VRM, they run more efficient with power delivery which can reduce wattage due to not having to up the amps to deliver the voltage requested. This is why VRM have an efficiency curve. If they run hotter, as they get closer to temp clamp down, you will be working them harder than running them at lower temps. That doesn't mean you want them to be freezing cold, either, which I'm sure Johnksss or others could talk more about from experience. But that can explain part of what he was seeing on wattage versus what others saw, even with his trying to match the uncore/ring frequency with other viewers that he collaborated with.

Also, did you notice the tease at the end of Wendell's video for the OC PCB or the 24 phase monster for the overclockable 28-core Intel Xeon? Just so much yes on that board!

 

The 9900K seems to have no gain over an identically 8700K in those games from what I've seen.

 

very impressive and super interesting especially for my particular usage case so i can see about 10 to 40 avg fps gains going from 7700K to 9900K. aside from that, wattages are in the mid 70s max with the 9900K during games, which is absolutely NO problem whatsoever for my machine, anything up to approx. 125W is thermally not constrained very nice findings!

also crazy to see how much gpu usage differs! more than seldom one can see 50-60% on the 7700K whereas the 9900K pegs the GPU at close to 100%

 

Keep in mind that was tested on an overclocked Vega 64 LC at 1080p low settings to push the CPU bottleneck to the forefront, so on a much slower 980M at higher settings you would see little to no difference between the CPUs since you are fundamentally GPU-limited.

 

oh im aware of that, no worries also planning on a gpu upgrade in combination with the 9900k besides, gaming is just a bonus for me, main focus is on cpu intensive applications

Sent from my Xiaomi Mi Max 2 (Oxygen) using Tapatalk

 

Well, this guy's drilling for "smoke"

9900K OC 5.2 With AVX2 load for 24hr on a Z390 Aorus extreme
Hi Cookie
Published on Oct 26, 2018
no air flow, custom loop, waster cooling by bitspower, the video is showing you, avx acting (I set avx offset to show it does run avx after p95 load), and how cool is, thanks to alva, I borrowed one IR camera to show the rig temperature. core i9 9900K is mind blowing fast in 8 core performance and allow you to push very high frequency for all core. if you want to max out CPU power, this board is the overkill design, wanna see more current on a i9-9900K?

Multiple cores @ 108c!! @293w...


9900K OC 5.2 With AVX2 load for 24hr on a Z390 Aorus extreme! VRM Cooling Off!
https://www.reddit.com/r/intel/comments/9rm1ir/9900k_oc_52_with_avx2_load_for_24hr_on_a_z390/

Intel engineer reveals details about Intel Core i9 9900K
Cleru Zinek
Published on Oct 25, 2018

Spoiler: Click to See Video

Thanks to @ Jan Prochnica :
http://forum.notebookreview.com/threads/the-random-video-thread.758474/page-165#post-10813560

 

Here is some balance for all of the talking heads in commercialized social media. Temps don't seem nearly as bad as some are portraying them to be, at least for this fellow. He has not delidded his 9900K (yet). Those interested may want to follow the thread to see more actual end user results.

9900k @ 5.0ghz asus maximus x code

@Papusan

 

If you look closely, he's only drawing 6.5w / 7.1w / 8.5w while those screen shots were taken so those temps are suspect...and he needs to adjust his Tjmax to 115c...that 9900k at higher OC than 5.0ghz will pass 100c easily.

 

When you were expecting your i9-9900K today
https://www.reddit.com/r/intel/comments/9pwzf8/when_you_were_expecting_your_i99900k_today/?sort=new


angelgr7 3 points 5 days ago
"Idk. It complete crap from amazon. I should have know not to order form them, after it was reported that their preorders for 2080 and 2080ti were delayed. Even though most other retailers were delivery on schedule. that why I’ve said that I might build a 2700x, at this point. But there different reports some Saying one to two weeks, other saying two months. And, considering that intel is struggling to make cpu right now. I think is most obvious that it will take 2 months at least. I’ll never preorder from amazon again, and truly considering not buying anything from them again. The only way to show them is with your wallet. By the way they blocked my from posting and comment on amazon."

drlaboo 1 point 5 days ago
"Same **** happening here in Malaysia. I got a hold of Maximus Hero XI COD BO but no 9th gen proc to be seen. Running the system with i3 8100 for the time being like WTF?!"

xray1337 2 points 5 days ago
"Nice setup! Also waiting... https://imgur.com/a/qiUU2Uk"

Kwerpi 1 point 4 days ago
"I am in the same situation except I have a full custom loop with flexible tubing filled and leaked tested just waiting for the 9900K."

SupaZT 6 points 3 hours ago*
"kill me fam. My Oct 8, 2:12 PM PST order is still nada. At least Amazon upgraded my shipping:

"...to help you with this I've successfully upgraded the shipped method from Two-Day to One-Day Shipping

I've successfully checked it with our internal inventories team and they've stated that this item will be back in stock by December 6 2018 - December 7 2018.

I'd like to tell you that as per policy and current stock availability the products are shipped to the customers on first in first out basis.

I provided you the maximum time that is being provided to us by the manufacturer as per the number of orders placed for this product on our website"
https://www.reddit.com/r/intel/comments/9rn4bt/amazon_i99900k_delivery_oct_30th/

So apparently this is a big thing now, lots of people recently recieved an email from Amazon saying their expected delivery isn't going to ship, and there is no estimated ship date, and people are suggesting going to Amazon chat to check up on your order...as the situation may cause your order to be canceled?!!:

Everyone Should Chat Support Amazon for Order Update
https://www.reddit.com/r/intel/comments/9ronsx/everyone_should_chat_support_amazon_for_order/

"People are being told Amazon when they pre-ordered they have issues authorizing peoples payments when they placed pre orders so they got pushed back 12 hours, and that caused the pre order to no longer be a pre order and basically lose your spot in que. So you may want to chat with them and make sure your order is a pre-order.
Amazon done screwed up."

Read that thread for more confusion...and this one:

Amazon.com Email Just Minute Ago Regarding my Pre-Order (Not Helpful!)
https://www.reddit.com/r/intel/comments/9rp0hm/amazoncom_email_just_minute_ago_regarding_my/

" Hello,
We're contacting you about your order for Intel Core i9-9900K Desktop Processor 8 Cores up to 5.0 GHz Turbo Unlocked LGA1151 300 Series 95W. We recently learned that we may miss your delivery promise and apologize for the inconvenience. You'll still receive the item and you can track the status of or make any changes to your order under Your Orders on Amazon.com.
https://www.amazon.com/gp/css/your-orders-access
We value your business and hope to see you again soon.
Sincerely,
Customer Service"

 

When you received your 9900K

 

https://imgur.com/a/GrSXQ1Z

When you also got your 9900K but it’s not installed yet due to date night.

 

No, not that at all. Nothing is suspect. The guy has no agenda, he's just a happy new 9900K owner.

What you may not realize is that the Windows software sensors are often not accurate and it is not uncommon for watts or voltage to not be displayed accurately in Windows, especially with a new product. When my 7960X is pulling over 1kW in an overclocked CPU benchmark, the Windows software shows like 1.5W-2.0W. When I had 8700K CPU-Z showed voltage at less than 1.000V at 5.2GHz. It was not until 2 or 3 BIOS updates and HWiNFO64 updates until everything looked right. It may take several months to get those sensors right in software, or maybe they will never be right since they are still not right on my current setup and it was released quite some time ago.

That is why I asked him to run more benchmarks. His Cinebench score seems about right and doesn't suggest any throttling or malfunction on face value.

It's also worth nothing that he is not using an AIO. He has a custom loop and 360MM radiator. I think that is a better solution for a 9900K than your typical AIO.

 

That's tomorrow night.

 

Glad you have your priorities straight. Keeping the lady in your life happy means you're going to be happy. After all, she gave you permission to buy it, right?

 

Not weighing in on the whether something is wrong or not (cannot see the images on my phone), rather on the software problems of reading voltages, frequency, etc.

When you see these issues, like what you saw when heavy OC session, not the behavior and report it as a bug. I did that with CPU-Z, noticing that it was reporting the wrong clock and frequency by 0.25GHz depending on which clock was set for the OS with HPET on or off and the bios switch for HPET (basically, I can, through those two switches control whether the ITSC, RTC, HPET is the default timer used on the TR platform). Only one timer, I think the ITSC, allowed for CPU-Z to read the correct value on the frequency. After I reported it in like May or June, it was fixed in the next CPU-Z release. I guess no one reported it or gave the detail and context to find and fix the problem before then. So sometimes waiting for a fix won't help, but reporting it with enough data and information to reproduce the bug does.

Edit: Sorry for the aside here, just wanted to give more info on software bugs in reading things. That also can help narrow down if your board has a bad sensor or some other hardware issue or off it is a software thing.

Sent from my SM-G900P using Tapatalk

 

Is that in your P870TM? If so, how much hotter is it than the 8086K? Temps harder to manage?

 

I haven't even ran a bench with it yet, but yes sir....This sucker is HOT, but then that was on factory settings and doing nothing. Just booting to see if it was going to boot.

 

nickname that sucker Buster Poindexter ... feelin hot hot hot
or --> Nelly ... hot in herre