seduce_me

Finally, managed to break 5k in Time Spy Extreme. Weirdly, other attempts to benches were no longer stable. GPU was boosted to 2,088MHz this time, and RAM primary timings were optimized to 13-15-15-33 at 3,466MHz. Secondary and tertiary timings also had changes but I can't remember much aside from follow the RAM OC guide in this forum. Very satisfied with the score right now. Benchmark: Time Spy Extreme 1.0 CPU: i7-8700K GPU: GTX 1080 Ti GPU Core: 2,088 MHz GPU Memory: 1,510 MHz Score: 4,948 3DMark Link: https://www.3dmark.com/spy/4421075 PCPartPicker Link: https://pcpartpicker.com/user/SydB/saved/gJwq4D GPU cooling was the bottleneck identified, less than 100% fan speed wouldn't be stable so I guess I'll need to watercool it as well. RAM secondary and tertiary timings could be further optimised but I'm done for now.

Synthetics wise looks the RTX2080 looks to be equal in performance to the GTX 1080 Ti. I have a decent OC on mine and get 4984 in Time Spy Extreme. Although the CPU score might affect it (i7-8700K vs i9-7900X) https://www.3dmark.com/3dm/28545564

CL16 at 2400MHz gives little room for OC to 3000MHz especially if it isn't Samsung B-die. Even if you manage to get it to 3000MHz at CL18 or CL19 the trade-off in the loosened timings are not worth it just to get 600MHz more out of your RAM. Personally, I have a 4400MHz CL19 rated kit and I have use 1.4V to hit a stable daily of 4000MHz CL16. That doesn't mean you can't try though, tightening the timings or a little bit more clock speed is always good. You can use stressapptest to verify the stability of your memory. For using stressapptest: - Download Linux Mint here -https://linuxmint.com/download.php - Format utility (ISO to USB) - rufus (open sourced) -https://rufus.akeo.ie Use the format utility to get a copy of linux mint on your thumb drive. Getting google stressapptest: - Open terminal - "sudo apt-get install stressapptest" - 1 hour stressapptest with settings recommended by Raja@Asus-> "stressapptest -M 14500 -W -s 3600" Up to 1.5V Dram should be safe for daily use but beyond that is not recommended.

- Sell i7-8086K, RX580 - Get used GTX1080 or 1080 Ti when prices drop due to RTX launch - Have fun Also I think the trade in for 1950X has ended already.

ASRock boards are great for the price and some of them have good OC performance on coffee lake chips.

No I'd stick to my Z370 Taichi for the beefy VRMs.

Assuming the current Z370 boards do not support the i9-9900K?

Can't really go wrong with so many options, just stay away from cheap OTL tube amps. It may sound enticing to pair with your HD6XXs with promises of a warm smooth sound. At best, it sounds limp, overly warm and a sense of buyer's remorse. At worst, the sound won't be the only thing that's warm. Your whole house will be on fire, sounds warm enough to me.

Lol no

What sound card are you using? Try plugging the cable into an appropriate 3.5mm output.

Valley 1080p Extreme HD Preset (was a bitch to screenshot in full screen, couldn't have GPU-Z and HWINFO64 open to screenshot in full screen but hopefully the other scores can validate this result) CPU - i7-8700K @ 5.3GHz, 1.45V LLC ExtremeGPU(s) - GTX1080Ti, OC'd to 1770MHz boost core clock, 1508 boost memory clock.Average FPS - 166.5Score - 6965 Valley 1440p 8xAA windowed CPU - i7-8700K @ 5.3GHz, 1.45V LLC ExtremeGPU(s) - GTX1080Ti, OC'd to 1770MHz boost core clock, 1508 boost memory clock.Average FPS - 97.7Score - 4086 out of space lol https://imgur.com/YL2HqW5 Valley 2160p 8xAA windowed CPU - i7-8700K @ 5.3GHz, 1.45V LLC ExtremeGPU(s) - GTX1080Ti, OC'd to 1770MHz boost core clock, 1508 boost memory clock.Average FPS - 48.7Score - 2037 https://drive.google.com/file/d/1yM4Hhmdn1XKaFxHXOZy2cclgnaAzrmLB/view?usp=sharing imgur didn't work

Here's my scores for both superposition and valley. ambient 18C, GPU is Aorus GTX1080Ti with stock air cooler, 5.3GHz i7-8700K water cooled with a single 240mm pump. I would like to be added to the leaderboard. Most of the benchmarks I couldn't get CPU-Z to display 5.3GHz constant so I had to use HWINFO64 for them. Hope it still works out. Superposition 1080p CPU - i7-8700K @ 5.3GHz, 1.45V LLC ExtremeGPU(s) - GTX1080Ti, OC'd to 1770MHz boost core clock, 1508 boost memory clockAverage FPS - 47.34Score - 6328 Superposition 4K Optimized preset CPU - i7-8700K @ 5.3GHz, 1.45V LLC ExtremeGPU(s) - GTX1080Ti, OC'd to 1770MHz boost core clock, 1508 boost memory clock.Average FPS - 78.00Score - 10428 Superposition 8K Optimized preset CPU - i7-8700K @ 5.3GHz, 1.45V LLC ExtremeGPU(s) - GTX1080Ti, OC'd to 1770MHz boost core clock, 1508 boost memory clock.Average FPS - 35.89Score - 4798 Not enough upload space to post valley scores, will do in the next post.

Looks great, have fun!!

Tightened the RAM sub-timings, all the tests have had an improvement in scores. Was bigger than expected, since I literally only changed 2 sub timings lol. I'm now top 93 in Fire Strike Ultra, top 142 in Fire Strike Extreme, top 207 for Fire Strike and top 150 for Time Spy Extreme for the same specs. Now I'm thinking the RAM is a bottleneck because I can't push any further without a watercooled GPU. Probably going to get a RAM with a more aggressive XMP profile to get reduced CAS timings and clock speed. Could I get top 100 in single GPU if I get a better binned GPU and RAM doing 4k+ MHz at C16 or less? Benchmark: Fire Strike 1.1 CPU: i7-8700K GPU: GTX 1080 Ti GPU Core: 2,076 MHz GPU Memory: 1,510 MHz Score: 26,330 3DMark Link: https://www.3dmark.com/fs/16321167 PCPartPicker Link: https://pcpartpicker.com/list/pmbn9J Benchmark: Fire Strike Extreme 1.1 CPU: i7-8700K GPU: GTX 1080 Ti GPU Core: 2,063 MHz GPU Memory: 1,510 MHz Score: 14,908 3DMark Link: https://www.3dmark.com/fs/16321362 PCPartPicker Link: https://pcpartpicker.com/list/pmbn9J Benchmark: Fire Strike Ultra 1.1 CPU: i7-8700K GPU: GTX 1080 Ti GPU Core: 2,063 MHz GPU Memory: 1,510 MHz Score: 7,899 3DMark Link: https://www.3dmark.com/fs/16322779 PCPartPicker Link: https://pcpartpicker.com/list/pmbn9J Unfortunately, I couldn't get Time Spy Extreme to get over 5k Benchmark: Time Spy 1.0 CPU: i7-8700K GPU: GTX 1080 Ti GPU Core: 2,063 MHz GPU Memory: 1,510 MHz Score: 10,633 3DMark Link: https://www.3dmark.com/spy/4377233 PCPartPicker Link: https://pcpartpicker.com/list/pmbn9J Benchmark: Time Spy Extreme 1.0 CPU: i7-8700K GPU: GTX 1080 Ti GPU Core: 2,063 MHz GPU Memory: 1,510 MHz Score: 4,948 3DMark Link: https://www.3dmark.com/spy/4340555 PCPartPicker Link: https://pcpartpicker.com/list/pmbn9J Also it's weird that the GPU core in the first test I run is always at 2,076 regardless the test but for subsequent tests it drops to 2,063 MHz. Planning to sell off the current memory I have and get a 16GB 4266MHz C19 kit and hopefully reach C15 4000MHz. I will watercool my GPU in the same loop soon with a single 240mm loop. Will it be sufficient or should i get another 120mm radiator?

Playing the devil's advocate here, but I'd say don't cancel your preorder. Even if it doesn't turn out to be your liking, you can sell it and regain a lot of the cost very quickly. Pretty sure there are many people who would want to get their hands on a like-new RTX2080Ti especially slightly below MSRP. If you like the performance but it runs hot, you can sell it or opt for a water cooled setup (EKWB blocks for FE are coming soon I think). All of this for what? You may be able to skip GPU shortages and artificially inflated prices. Your call, OP.

Weird, personally I find the key rebound on my HHKB to be faster than the most common Cherry MX variants. I do bottom out so that could explain why or maybe it's the HHKB?

Mouse for what kind of games? Different games have different kinds of mice that perform better. Also what is your hand size?

Topre. Good feeling of O N E N E S S with cup rubber. In all seriousness, I really like the upstroke speed which lets me type faster and with less effort. Sounds great too.

Hero sensor matches the PMW3366 used, no upgrades there except for lower cost. The omron switches with longer lifespan are a nice touch and flexible cable should be better, though 121 grams without any weights or the cable is still waaaaaaaaay too heavy to be an appreciable difference. I'll pass.

i5-8400. More cores.

In short, better VRM implementations are more stable in voltage so in terms of overclocking a better VRM can translate to lower Vcore for the same clock speed hence more stable. Also, better VRM implementations are able to handle the higher power draw of overclocking that low-end ones can't. Personally, my i7-8700K at 5.3GHz draws about 120W on average at full load. Here's a more technical and in-depth article that explains the benefits and what makes a good VRM: http://cxzoid.blogspot.com/2015/04/what-makes-good-vrm.html

I updated my post, if you are not interested in overclocking at all the B360 board in the OP would suffice.

I would spend 10 euros more on the ASRock Fatal1ty K6 if you are using the same website: https://tweakers.net/pricewatch/1076833/asrock-fatal1ty-z370-gaming-k6.html But you mentioned you are not interested in overclocking so a H370 or B360 motherboard would suffice. The VRM shouldn't matter in these cases since there is no overclock.

If the miner undervolted his GPUs and always kept the GPUs running constant and well maintained it would actually last quite long. The problem is that miners kill the fans on the GPU so if you can get those replaced it would be quite reliable I'd reckon.

Wireless mice are pretty good these days. See G900, G305/G304. I had a few mice in this order: Razer Taipan > G502 > G900 > G Pro > G100S > G305 > G100S. If you ask me why I didn't stick to the wireless mice, I just liked the shape of the G100S way more than the rest. Which is a pity, because if they made a G100S with the same wireless technology as the G305 it would be the perfect mouse for me.