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

yep, well i dont mind paying extra for higher binned. this time they are going to remove iGPU, so that 9900KF better be much cheaper than 9900k and if it isn't we know intel is up to no gud again. (or business gud)

 

Although that was cheaper than buying a silicon lottery binned 8700K, to be honest, excluding the delid kit. 100% of 8086K's would reach 5 ghz on 1.35v or lower.

 

Still won't help the chips clock better.
Didn't someone say long ago something about binning?

Like early shipments of a new process (tick or tock?) usually had good overclocking results (Intel shipping out more good batches to give themselves a good name? who knows. Hey at least they aren't shipping out $1200 packages with a free copy of Space Invaders!)

then as time went on, the chips would often overclock worse in much later shipments.
Then close to the end of life of the product run, the chips overclocked better again because the process got more refined.

I don't know if that was true or just heresay.
But a 9900K isn't even a tick or tock. It's just an 8086K with moar corez.
(disable 2 cores on a 9900K and watch 100% of them suddenly reach 5 ghz. Someone prove me wrong).

 

not sure, we can only hope it'll be binned better a true 14nm+++ or ++++ there can be only so many +'s

10 core comet lake in clevo i can smell it

 

You talk about Intel. I talk about out from the Stores (market place). More chips out will push the prices down. The stores can't charge what they want if there is enough chips out there. And chips without working iGPU has to be cheaper out from Intel.

You can not compare it with a jubilee's chips.

 

intel might as well raise that MRSP. 9900k right now at 500-550 ish, 9900kf im hoping to see it at 400 but im guessing it'll be around 450-500. $50 for iGPU mmmmm

 

Have never seen Intel increase recommended prices for older "current" chips.

 

i hope thats the case cause we dont want to spend more. also with ryzen 2 around that time i dont think they can, just have to wait and see

 

The 9900K wiping the floor with the 4.2Ghz overclocked R7 2700. In some games the 2700 at 4.2Ghz was marginally better than a very old overclocked 2600K.

 

2600x feel so long ago and sluggish now. hoping to see some zen 2 performance though can't get enough of competition.

 

Something went wrong with the small watercooled PC...9900K
JayzTwoCents
Published on Dec 3, 2018
Had to tear the small watercooled PC apart again... watch to find out why...

 

Top100 9900k in the Antec Torque ft. Gigabyte 2080ti
Level1Techs
Published on Dec 6, 2018

Spoiler: Details...

Case: https://www.antec.com/product/case/to...

CPU: http://www.amazon.com/dp/B005404P9I/?...

GPU: http://www.amazon.com/dp/B07GJKR7RW/t...

Motherboard: http://www.amazon.com/dp/B07HS4PQWK/t...

RAM: http://www.amazon.com/dp/B01MYCKM2Y/t...

PSU: http://www.amazon.com/dp/B0009P7J0K/t...

 

I know I deleted some valuable info in here, however it was too intertwined with the other argumentative posts that needed to be deleted. Let's move on.

Charles

​

 

https://www.anandtech.com/show/13683/intel-euvenabled-7nm-process-tech-is-on-track

Intel 7nm could be here sooner than we though. Maybe 10nm will be a very short lived product?

 

the amount of butt hurt doesnt seem to stop with amd vs intel. bring on zen 2 already can't wait

 

At least on 9900K, the CPU's "VID" table doesn't affect static (manual) voltage overrides like they do on all of these laptops. (eVGA seems to be an exception).
And on some of the new motherboards with the IR 35201 buck controller, you can read the EXACT CPU voltage that it's getting, unaffected by power plane impedance.

https://www.infineon.com/dgdl/Infin...N.pdf?fileId=5546d462576f347501579c95d19772b5

(called VR VOUT in HWInfo64).

 

wait! so slapping a 9900k into a laptop would mean you could actually use REAL static voltage without VID override, rendering static basically unuseable?

 

Intel i5-9600K Review vs. R7 2700, R5 2600, i7-8700K, et al.
Gamers Nexus
Published on Nov 26, 2018
We benchmarked the Intel i5-9600K for gaming, production, and overclocking.

Spoiler: Details...

Article: https://www.gamersnexus.net/hwreviews/3407-intel-i5-9600k-cpu-review-vs-2700-2600-8700k
Intel's i5-9600K is the next in line with the 9900K and 9700K (the latter of which we haven't worked with yet), serving a more mid-range market at ~$250. The 9600K runs 6C/6T, similar to the 8600K, but has been updated with solder and higher native clock speeds. Overclocking the 9600K is also interesting, something we do in the review video above.

The Intel Core i5 9600k Review
Timmy Joe PC Tech
Published on Nov 12, 2018
With Intel struggling to produce high end chips my only option was the Core i5 9600k and we quickly find it's a terrible value vs the Ryzen 5 2600 and barely justifies even existing.

Spoiler: Click to See Video

Where is Intel's i5-9400...? AMD's 7nm Ryzen 3000 Series is Coming!
Tech YES City
Published on Dec 6, 2018
So upon recently doing up budget orientated videos for you guys, I noticed that Intel hasn't really been hitting the list, as not only has the Ryzen 5 2600 come down to extremely good levels, but also the i5-8400 has gone UP in price...!? To make things worse for Intel the i5-9400 hasn't been released, nor has it been even rumored at this stage.... So with all the upsides for AMD going into 2019, with increased sales dominating 2 to 1 and the upcoming 7nm that is rumored to hit 5GHz AND have more cores, it makes the roadmap that of a very successful one for AMD.
Articles Cited: AdoredTV: https://www.youtube.com/watch?v=PCdsT...
AMD Outselling Intel in Germany: https://wccftech.com/amd-dominates-re...
Intel 10 Core Rumors: https://www.forbes.com/sites/antonyle...

Spoiler: Click to See Video

 

No. Nothing to do with the CPU and everything to do with having a properly engineered desktop motherboard BIOS!

 

then why were u referring to the 9900K and not specific mobos?

 

From his post "At least on 9900K, the CPU's "VID" table doesn't affect static (manual) voltage overrides like they do on all of these laptops."

 

derp, thats what happens when you actually focus on work instead of reading the forums properly

 

Wrong thread

http://forum.notebookreview.com/thr...ga-polaris-gpus.799348/page-526#post-10831002

 

9900K in MSI 17 inch workstation (with C246 server chipset, not Z370 or Z390 because this is their workstation model that takes Xeons and Quadros)

https://www.notebookcheck.net/Euroc...kstation-Review.373721.0.html#toc-performance

Looks like it has a 80W CPU power limit throttle leading to ~3.9ghz on prime95, its cinebench r15 is down around 1600... so given the factory underclock... temps are good.... hooray?

 

As I said many times, this is an EC problem, because the EC isn't coded to recognize LGA CPU's. So the EC overrides any Bios power limit manual settings and enforces CPU MSR TDP values. This can be by passed by decreasing IMON SLOPE and setting IMON OFFSET to negative offset of 31999 or 63999.

Exact same problem was present in the MSI F5.

 

8 cores at only 80w, yummy

 

Below base clock in AVX loads, yummy indeed.

 

Here is the case my other computer was using previously (Core V9). Moved my 2 x 240 mm radiators over and added a 360 mm. Picked up a 9900K from Newegg as well and below are a few photos and a few benches at 5.0, 1.345 (showing highest 1.353 in HWInfo) which was the lowest stable i could find at 5.0 without using AVX offset. I ran some tests with Prime95 as well but the temperatures were just too high for me to run for extended period of time so i ran Aida64 blended test for about 7 hours. Was able to boot into 5.2 at around 1.42 which failed after a short prime95 run but 1.43 ran for about 10 minutes although again the temps were just too high. Maybe some more tests at higher clocks in the future.

Edit: a little more info about the Cinebench runs. The top score was at 5.2/4.7 cache (vcore not dialed in), the second score was at 5.0/4.7, the third score was at stock, the fourth score was the 9900K at 5.0/4.7 with hyper-threading disabled and the fifth score was 9700K at 5.0/4.7.

 

Cool, so you've got x1 360mm radiator and then x2 240mm radiators to cool down your PC - but is it the 360mm radiator for the CPU and then the x2 240mm radiators for the GPU, or which radiators cool which components? Or they're all joined together in the same loop? If they're all joined together in one loop, then whats the flow direction & 'sequence'?

 

I do not have 2 pumps yet, had thought about that in the future. They are all in the same loop. Right now it goes from the pump, to the video card, to the processor into the 360mm rad on the top, into the 240mm rad on the top, into the 240mm rad on the front and then back to the pump. Was trying to save on hose length. There may be a better orientation but not sure it would make much of a difference. I tried looking it up and saw a lot of people saying the orientation did not matter all that much but rather the fans either being intake or exhaust. I was considering putting the 360 on the side as intake; however, after looking at it i could not figure out how to mount it and would need to mod the case or buy a specific bracket i guess. So as of now i have a 240mm as intake in the front, one 140mm fan in the back as exhaust and the 240mm and 360mm on the top as exhaust. May look into other fans but temps on the board seem to be fine.

Edit: let me know if you have experience if there truly is a better orientation, next time i flush it i will give it a try.

 

I don't have any direct experience with water cooling, but I like to understand systems and have a scientific background that involves some heat transfer and 'flow', so I just like to visualise & understand these things. You explained that layout well, so I can visualise that. At the moment you have it set up so that the coolest water is entering the GPU, and I would do it that way around too, rather than putting the CPU first, because to me I think cooling your GPU to lowest possible temperatures will give you the largest practical gains - will give you max clock boost potential on the GPU as well as might help gain a larger overclock from it. The CPU might well be harder to cool than the GPU, but the CPU doesn't drop it's clocks with increased heat like the GPU would, so not such a crucial thing to have the CPU running cooler. You might be able to get a higher overclock if your CPU is running cooler, but that's a big maybe, and on a practical level (both gaming & productivity) the difference in overclock would be very small, probably 100Mhz difference at the maximum if at all, and that just will not affect your gaming performance at all really because 9900K is overkill/ideal for games as it is.

I would think 2 seperate loops & 2 pumps would provide the best cooling performance, but by how much I do not know - the mechanism for increased cooling performance would be more efficient/higher flow rates around the loop as well as larger volumes of cooling water (2 reservoirs) so more heat capacity in terms of taking a longer time to heat up the larger volume of water (so would absorb blips of high power output better). If you do split it into two loops though you'd have to decide which gets the 360mm radiator and which one get the x2 240mm radiators - I'd probably run it so that the GPU didn't have any boost clock throttling, so I'd choose the most efficient one for the GPU cooling - but on the flip side of that if both loops run the GPU below boost clock throttling temperature anyway then I would choose the most efficient loop for the CPU.

 

That was my thought as well, when i start gaming this week with the 9900K overclocked i will see how much my GPU heats up but with the 9700K in the same loop setup at 5.0 ghz i was seeing around 40-43 C max on the GPU with it overclocked and around 62 max CPU temps. I would like to keep the GPU under 50C if possible.

 

VR VOUT is the most accurate reading for the cpu vcore.
https://www.overclock.net/forum/5-i...load-voltage-main-1-watch-2.html#post27736104

 

Thanks, I will take a look at that tonight. I thought I had the correct Vcore marked from what someone told me on the gigabyte site.

 

The vcore you marked is from sensor #2, which keeps the idle and load vcore about the same when using "Turbo".
Elmor told me that that vcore is probably measured from the MLCC caps behind the socket. The Multimeter read points on the Aorus matches what that sensor says, but it is still affected by ground plane impedance.

https://www.overclock.net/forum/27686004-post2664.html

vcore sensor #1 would be the light blue line, sensor #2 would be the purple line and VR VOUT would be olive.

 

Interesting, a little complicated with the additional sensors.

 

The Worst CPU & GPU Purchases of 2018
Hardware Unboxed
Published on Dec 13, 2018
Skylake-X starts @ 06:20, i9-9900k @ 14:22

Spoiler: Click to See Video

Core i9-9900K Upgrade, Can We Cool This Hot CPU? Corsair H115i Platinum Tested!
Hardware Unboxed
Published on Dec 12, 2018

Spoiler: Click to See Video

This is what you need to keep the 9900K at 95° C!
submitted 2 days ago by anmols4096
https://www.reddit.com/r/pcmasterra...s_is_what_you_need_to_keep_the_9900k_at_95_c/

Spoiler: New 9900k air-cooling solution...

https://www.reddit.com/r/AyyMD/comments/a5ibry/pc_master_race_knows_whats_up/

 

Nice, where do we pick up one of those Beafy CPU Air Coolers?

 

Right?! I saw that and thought - hmmm, I've got a couple of cases I could probably fit that in...

It looks like it's a photoshop'd special, so if only someone gets inspired by the image to made one in real life will we get to buy one.

 

Here is some testing at 5.2/4.7 - 1.435 Bios Vcore - CPU Loadline (Turbo):



Aida64 Blend - Just 15 minutes.



Intel Burn Test (Maximum).



Battlefield V (45 Minutes).



COD Black Ops 4 (45 Minutes).

 

How come Intel Burn Test had some of the lowest reported CPU Package Powers out of all your tests, I thought it was supposed to be pretty much the most demanding test?

I also noticed that you're over 1.4V VCore, are you sure you wanna run that as your daily overclock? I got a small amount of CPU degredation on my Skylake i7-6700K by running it at 1.4V for over a year, and that was only at operating temps up to 65 degC max, I think you are really quite likely to see some CPU degredation within a year if you run over 1.4V, especially given the relatively high temps your seeing there. I think you're safe with 1.37V as a maximum VCore, and that's how I've chosen to run my CPU now (lowered my overclock and reduced VCore a few months ago).

 

That vcore is a bit too high.
You can score over 2250 at 5.1 ghz. (I managed 2286 CB once).
Set system agent voltage to 1.25v. Vccio to 1.15v.
Then set Command Rate to 1T, tRFC to 270 and tREFI to 32767.
This should improve your scores and bandwidth.

 

No I am running 5.0 as a daily driver I just wanted to test the temperatures at 5.2. I am not sure about intel burn test, first time I have run it in a long time. Also, odd to me that battlefield V saw higher temperatures than all the tests.
Edit: looked at the temps again. Was not the highest but that game is demanding.

 

Thanks, I will give those settings a try. I will probably stick with 5.0 for gaming. Temps with my video card in the loop are as high as I would like to see them. 61-70 C on the cpu depending on the game. GPU never seems to go over 45C.

 

That's better for 24/7, what VCore you at for 5.0Ghz?

 

Lowest I could hit on my chip was 1.345. I can go lower with avx offset like silicon lottery shows on there 5.0 chips but I want to be stable without a avx offset.

 

Well that's just fine, I don't think you'll be seeing any silicon degredation at your 5.0Ghz.

 

This chip was actually a chip I got from Newegg and Rockit Cool delided it and put some CLP on it for me with a copper IHS. Live right down the street from the business. Did not really do any testing before to see the difference in temps but I know they ran some tests on there own. They only charged me for the copper IHS which was really nice, but I ended up giving them some extra cash for the work. I did do some gaming previous at stock though and looked to be around 8C at stock.

 

Typo? 8C at 5GHz on the CPU?

 

The temp difference with delidding and copper IHS vs. stock Intel chips.