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

Very nice results with that Gen 3 CPU.

 

Core i7 12700KF reviewed. This is looking to be the sweet spot of the Alder Lake lineup as you get most of the performance of the i9 but at a lower cost with much better thermals. Gaming is pretty much a toss-up between it and the Ryzen 7 5800X, but Intel absolutely dominates in productivity, even with DDR4 memory. AMD is going to have to cut its pricing if it wants to be competitive until the 6000 series launches.

 

Have found exactly what he did. I am able to overclock and undervolt at the same time. Seems these chips are running high voltage out of box, similar to how the 8700K ran far higher voltages than necessary out of box.

 

Looks like except TPU nobody even bothered to test these CPUs with E cores disabled in gaming, yes they have some variance in benches which I mentioned in my first post about how 10th gen got kicked down suddenly on the same test rig anyways it's a damn shame nobody wants to cover that part at all. They do not do anything at all in gaming, but all these youtubers are simply calling omg these cores are real deal and the massive boost is real and Gaming King !!



Steve literally bashes 11th gen in every review just watch his i5 review below, constant bashing on 11th gen, yes it needs but doesn't care anything past that to think like how Intel purposefully sabotaged LGA1200 with stop gap milking scam update while keeping the new Golden Cove P cores making real jump in ST and MT performance massive boost is all coming from these P cores only, literally Intel ploy to make 10th - 11th - 12th where the bar graphs show 12th make a massive leap, but in reality single digit % growth overall picture in gaming vs Ryzen 5000 vs 11th gen or even 10th gen. Nobody is unfortunately caring much on the E core P core drama either, the DRM Denuvo means one must disable the SKL cores in the BIOS else games do not run.



Why do you guys think AC Odyssey magically vanished from all the damn benchmarks reviews ? It doesn't run on Alder Lake. That's why, only way is to disable the E cores.

On the technicality side and really comparing and making what are those E Cores only Anandtech did the legitimate dive in and tell us what they are for and how much of the real lead is all coming from those big P cores on Golden Cove design and not the SKL cores. Literally in the conclusion. Also Ian dedicated a page explaining how the BS focusing of the E cores and P cores are annoying issue with Windows 10 / 11 relationship.

Anyways I digress, just can't unsee what Intel did to the LGA1200 and now god knows what plan they have for Z790 and DDR4 EOL.

 

Haven't tested that game myself but Tech Deals had no issue running AC: Odyssey with his ADL CPU.



It's very early days for ADL, remember when games wouldn't run on Zen 2 when it launched? No disabling cores made it run, it just didn't work for a month? Intel has stated they are working on getting DRM games updated and they have put a temporary work around via scroll lock to disable/suspend E Cores.

https://www.intel.com/content/www/u...0088259/processors/intel-core-processors.html

 

This has to be the first time in who knows how long that the scroll lock is serving a purpose, LOL.

 

Gonna stick with the 11900K for now at least, it’s tuned so well too. I have the opportunity to buy a 12900K. But I think I’m good for now. I literally just bought my current platform a little bit over over month ago. Which was released not very long before that. I’ve gotta wait at least 60 days between $600 dollar motherboard purchases and $600 dollar CPU purchases.

 

I'm sure future BIOS adjustments and mc updates will fine tune pull and performance a bit. 11th in ABT mode was defaulting to 1.45v when many CPUs didn't need anywhere near that to do 5.1 all core.

There's just no getting around it. Intel clearly won this round overall. It's just a fact. Doesn't mean AMD chips suddenly suck (just like I never thought Intel chips sucked before too).

12600k vs 5600x is pretty scary beat down overall outside of gaming. Gaming? No need to upgrade across all resolutions but the same could be said even running a nicely tuned 9900k, 10900k and 11900k especially at >=1440p.



12700k vs 5800x is the same slaughter outside of gaming where if you're gaming no need to upgrade. Really the only saving grace for AMD at this moment is the 5950x and that is in very heavily threaded workloads and even then they trade blows.



12700k vs 5900x is the same as the above in benchmarks and productivity just not nearly as destructive. Yet again, if you're gaming really no need to upgrade.




12900k vs 5950x at least it gets interesting with both going back and forth but yet again gaming? Really no need to upgrade.

I agree with @saturnotaku that the 12700k is a sweet spot delivering the full P-core performance. I think the 12600k is going to be the new go to gamer sweet spot especially when B660 boards arrive. Absolutely no reason to go with a 5800x or 5600x from this point on.

It's a pretty bad beat down except for the 5950x and that is only in full multicore loads and even then the 12900k is always right there and winning plenty. As the core load lessens even the 12700k sneaks in there and starts to take the lead over the 5950x. The 12700k even beats the 5900x as predicted. Intel is firing on all cylinders with the 12th gen and I say that STILL not liking the hybrid design instead of delivering a true 12 core / 24 thread and 16 core / 32 thread CPU and making the desktop computing scene overly complicated in a chase for the win with this hybrid nonsense. Like RL, AL is hot and heavy versus AMD but a win is a win.

AMD will have to lower prices but more importantly sooner is better for 3D cache refresh. At least AMD lit a fire under Intel to move quicker past skylake and 14nm++++++(++++++++++++)

 

Zero reason to upgrade. The argument could be made for numerous older chips too. Enthusiasts upgrade because it is what they do. I picked up a budget board and 12700k and even that was ~$700. A mid to high end upgrade (12900k + better motherboard) will run you $1000-1400 and that is before picking up DDR5.

A full bore top end upgrade with AL is very pricey.

I'll be running Windows 11 and Ubuntu exclusively on the test rig.

Even if I end up slotting in the 12700k to my main rig, I'll still hold onto my "cool hand luke" 11900k for the SM (or a KM?). I still haven't opened my factory sealed Z590 Dark yet while I test the 12th gen.

Speaking of which:

 

https://hwbot.org/submission/4847541_papusan_7_zip_core_i7_3770k_37613_mips


https://hwbot.org/benchmark/wprime_...Id=processor_2493&cores=4#start=0#interval=20


https://hwbot.org/benchmark/y-crunc...Id=processor_2493&cores=4#start=0#interval=20


Not small and pretty as Dell's Garbage but functional. So far with 12 fans

 

You crazy SOB! You actually did it! .

Looking forward to your testing. My second 12900K just arrived but won't be able to test it until Sunday most likely due to weekend wedding festivities for family member. I found that 12th gen has 2 SP scores. SP for P and E cores are separate and then it generates an overall SP score on main menu. My SP are 95 for P cores and 70 for E cores. I need to see more results to see how this shakes out.

Also doing a bit of efficiency testing. I am down to 1.175v on stock clocks with LLC4. This runs at 4.9Ghz P cores all core and 3.7Ghz E cores all core.

Package power draw in CB23 at those clocks/voltage is just 171w . That is a huge power savings over default. CB23 score was 27K. So not far off from stock voltage curve. I need to do a lot more testing to find bottom and then stability. But initial findings seems ADL is hugely overvolted by default.

 

This definitely needs a delid. Very hot without the chiller for some reason. Probably the nutty high voltage.

Are any of you with 11900K seeing the WHEA "PCI/PCIe Bus errors" in HWINFO64? I am not sure what is causing it. Playing around with memory doesn't seem to affect it. Bugs me that I had similar crap with Ryzen. I might just put the 11900K in the Strix Z490 board and put the 10900KF in the Z590 Dark. This runs good, but I can tell it is 2 cores short of a full load in some workloads. I am not used to it yet and haven't decided. I do not like the Gear 1/2 crap with the memory. I will decided after I delid and go bare die because it's not a fair test right now. I may change my mind. Too early to tell.

https://hwbot.org/submission/484763...i_core_with_benchmate_core_i9_11900k_16972_cb


Here are the errors I am referring to. I am trying to make them go away, but none of the settings I have tweaked have eliminated them.

 

Check mounting pressure. I think I had that the first day I got my 11900K and it went away with a BIOS update from Asus and or remount. Of course my first Hero board died a sad death and after swap with Asus all was good. I had thought it was my CPU, turned out to have been my freaking Z590 Hero. Asus was easy to deal with though and from submission of RMA online to replacement in mail was like 10 days.

But other than that, my 11900K was rock solid platform.

 

There should not be any errors on the 11th gen platform afaik... I hear some folks on OCN mention when AVX loads hits too high clocks on 5.3Ghz+ on 10th gen the CML starts having cache error but it can be mitigated by reducing the Cache clock ratio or upping the voltage. I think this is the exact reason why Intel made 11th gen not to have high cache ratio OC and instead maxed at 45x multiplier or such, Luumi was saying that 11th gen Cache Ratio is not very high for OC and it's lower.

I just remembered that I came across something like this on EVGA forum a while back just searched for it and here's this thread. Unfortunately it's hard to diagnose..some mention it could be the BIOS, or the Memory or CPU or Mobo.

Can you try to run the Memory on Gear 1 3200MHz, I see it's Gear 2 speed. I think aim for Gear 1 only, Gear 2 is waste of time and not worth, unless we push past 5000MHz+ which EVGA boards cannot do despite their marketing about 5333MHz. I bet switching to PCIe3.0 will fix it as it's the same on reddit thread linked there and a a few folks as well. But this kind of thing should not happen on 11th gen...wtf is really going on.

 

I think I answered my own question, and now I wonder if the same issue was occurring with the X570 platform's insane issue with hundreds and even thousands of the similar CPU bus WHEA errors.

I manually changed all of the PCIe slots to Gen 3 and *BOOM* now there are no more WHEA errors. I am running a Gen 3 NVMe SSD in the slot capable of Gen 4 and I was doing the same thing on the X570 mobo. I suspect that TONS of people are doing the same and I wonder if the people that do not experience the WHEA errors are not populating the Gen 4 M.2 slot, or only using a Gen 4 NVMe in that slot. This produces no malfunction that anyone can see, so I wonder if the errors are because the slot is trying to push more data through the SSD than it can handle so it results in a bottleneck that requires error correction.

@Falkentyne as an FYI for your frequent posting in ASUS forum. Something for both Intel and AMD users to investigate regarding the WHEA errors.

 

But NVMe and PCIe are backwards compatible like SATA standard. They should not have any issue like that...the first NVMe PCIe4.0x4 SSD slot is exclusive for RKL platform only and it has direct CPU lanes for SSD if we plug in 3.0 SSD the lanes will take up full 4.0x4 but the SSD will run only at 3.0x4 speed only. If we put a 10th gen 10900K then if we use that top NVMe SSD slot, it won't work simply because CML doesn't have those extra 4 lanes, no 3.0 no 4.0 no lanes should be there.

If we put a 4.0 SSD in M.2 2nd slot on Z590 it will only run in 3.0x4 (For CML it will eat up the GPU lanes if we do that to 3.0x8. For RKL it will eat the GPU lanes as well which is expected but the GPU will run at 4.0x8 which is 3.0x16) speed as usual with perfect backwards compat. It should be HW level just like USB.

RKL is the first Intel CPU to have full 4.0 PCIe I/O. The first 4 go to SSD, then the usual fare of x16 to the GPU if we split then x8 and x8. Z590 chipset gets DMI 3.0x8 lanes (Matches X570 which is at 4.0x4) if we use RKL, otherwise the DMI is 3.0x4 only if we use CML.

Anyways good to hear it doesn't happen on 3.0, but I think you should investigate...imho.

X570 WHEA issue however is very prevalent, right. At-least from what I was able to gather. So many of them had that esp when pushing for FCLK and IF 1:1 ratio speeds, some folks even get it at XMP / DOCP DDR4 3600MHz speed on Ryzen platform, below are reddit but you know the OCN pages already.

Code:

https://old.reddit.com/r/Amd/comments/lnmet0/an_update_on_usb_connectivity_with_500_series/?sort=new

https://old.reddit.com/r/Amd/comments/pujl1q/explanation_and_investigation_on_b550_chipset_and/

https://old.reddit.com/r/Amd/comments/oru7fi/with_agesa_1203_b_do_you_still_have_usb_issues/

Honestly **** Intel and AMD.

 

Yes, similar issues and errors, but originating from a slightly different source. There are some YouTube videos mentioning that setting Gen 3 for all PCIe slots stops or slows down the Ryzen WHEA errors associated with the FCLK higher than 1800. Also similar in circumstance is that the WHEA errors on both platforms produce no evidence of malfunction. It is as if the WHEA errors are essentially meaningless nonsense and the only harm evident is the extreme annoyance that their existence causes the system owners.

I do not know that my analysis of the underlying issue is precise, but there is obviously a connection there since manually setting Gen 3 fixes it. I am not going to waste money on a Gen 4 NVMe SSD to test whether installing one corrects the errors. I might remove an SSD from that slot temporarily to see if leaving it vacant and setting the speeds back to Gen 4 results in no errors.

If I end up deciding to move the 10900KF and install the 11900K in the Z490 system instead for the higher core count on my benching rig that will end up being a moot point. There was no measurable benefit to having the 3090 running at Gen 4 speeds in the Ryzen setup. Benchmark scores were unchanged.

Edit: I personally believe the WHEA errors (in both circumstances) are actually Windows glitching out and being flakey; not an actual hardware malfunction. I think the speed at which data is moving with some of the newest hardware is more than Windoze can handle without bugging out. I also think that speed is wasted calories on something that might matter someday, but represents little more than hype today.

 

https://hwbot.org/submission/4847713_papusan_cinebench___r20_core_i7_3770k_1997_cb


https://hwbot.org/submission/4847708_papusan_y_cruncher___pi_1b_core_i7_3770k_3min_6sec_769ms


https://hwbot.org/submission/4847719_papusan_7_zip_core_i7_3770k_38206_mips?recalculate=true


https://hwbot.org/submission/4847735_papusan_cinebench___r11.5_with_benchmate_core_i7_3770k_10.32_cb

 

When delidding the 12900k people will need to take in count the little smd cap on the substrate, which is the same height as the die if they are going direct die route. However, I bet a copper shim would solve this issue though.

 

When retards like this work for places like Micro$lop, I think having something that partially functions is miraculous. I wish we could say the same for their brains, LOL. Unbelievable idiocy. Thanks Micro$lop.

https://twitter.com/Lukewearechange/status/1456311555134459904?s=20
https://twitter.com/Lukewearechange/status/1456364384255623169?s=20

 

https://hwbot.org/submission/4847752_papusan_cinebench___r20_core_i7_3770k_2011_cb?recalculate=true


https://hwbot.org/submission/4847770_papusan_cinebench___2003_core_i7_3770k_4496_cb?recalculate=true


https://hwbot.org/submission/4847746_papusan_cinebench___r15_with_benchmate_core_i7_3770k_949_cb


https://hwbot.org/submission/484779...s_core_i7_3770k_13251_points?recalculate=true

 

They're burning all of their calories making toys with smartphone e-cores to titillate the kiddos rather than using all big-boy cores. Is it better? Yup. Is it as good as having all p-cores? Nope. But, the kiddies don't know what they don't know. That said, it does make me happy to see such a hard gut punch to Team Red. They've already become too complacent and need a good, hard slap in the face or they will slip back into a state of irrelevant obscurity again.

 

I only encountered errors running my 5800x + MSI X570 Tomahawk when pushing fclk > 3733. I ran both a gen 3 (970 Pro) and gen 4 (SN850) M.2 drive and neither scenario produced Whea errors. It was only when pushing the fclk 1:1. Manual overclocking would bork out when pushed in a proper way but also never produced WHEA errors.

I never experienced WHEA errors on any of my 11900k motherboards (MSI Z590-A Pro, Gigabyte Z590 Aorus Pro AX, Asus Prime B560M) in any fashion. If I pushed Gear1 too hard it would error up in TM5. I only ran SN850s on those though.

I have parts strewn all over from trying to isolate the dips in CPU frequency running the 11900k and TimeSpy (3x PSUs, 3x GPUs, 2x Memory kits, 2x M.2's) and the problem still persisted. Finally swapped the 11900k into the wife's system which has my first and old MSI Z590-A Pro motherboard and just like that running Timespy with her 3070 and the CPU line was rock solid like it was supposed to be. Swapped her i3-10105 back into my Gigabyte and it wasn't dipping but the 11900k still was. Down clocked my 11900k all the way to ~4.2ghz all core and suddenly the dips went away so it looks like the GigaFail motherboard is weak sauce.

I have so much reference data from my 11700k + MSI Z590-A Pro it will be good to draw up a good comparison between the two along with my 11900k's. I'm hoping I can resist the urge to start binning a small selection of 12700k's or 12900k's. I told myself last time don't do it and proceeded to test 5 of them.

I wonder if the mid to upper end MSI boards will differentiate silicon quality between e and p cores with their "CPU Force" rating (aka SP for MSI)? I'll have to keep an eye on the 12th gen forums over on ocn to see how the binning pans out and the vcore ranges. If the range is relatively the same as 11th then you have a P-Core winner (again).






So I'll re-assemble the test case/bench and get it installed this weekend and take a looksy.

 

https://hwbot.org/submission/484780...0m_core_i7_3770k_22sec_795ms?recalculate=true


Intel created a Hybrid monster with the added small baby cores. The bot in fire. People will loose loads of Global points more than ever before.


A 16 cores 12900K can get points in the 8 core processor ranking. The same also for the 12700K

Sudden loss of points Hwbot.org/forum

Then we have exactly the same (Hybrid mess) for 10 core 12600K. Can be benched as 6 cores and grab Global points from the 6 core ranking. Not fun with Ryzen 5800X on the bot nowadays, LOOL

And you should buy two Intel systems for benching on the bot. One with DDR5 and one with DDR4

 

I can confirm I can’t get 10Th Gen to run past 5.3Ghz on Z590 Dark. It down clocks once load hits. And it’s different every time.

 

Intel shenanigans at play 100% there's no doubt how this scum operates. As much as I respect Intel for giving the world x86 I equally hate them lol.

Hwbot 2xpts with magical ADL CPUs unfortunate for all the folks who work hard for their passion to get such treatment. Then we have Windows 11 itself literally feels like created for Intel, add the L3 AMD fallout, VBS etc issues and all reviewers suddenly forget what is E core and P core simply looks at Bar charts except a very few outlets. Today everyone says DDR4 is the way to go, wait for next refresh, everyone will say sorry mate your PC is outdated as Intel wants DDR5 only. Dumb consumer gets shafted.

Sorry, I actually meant DRAM frequency past 5000 to 5300MHz Gear 2. About the 5.3Ghz throttling you are mentioning I think someone had this issue on EVGA forum but they were able to fix with Throttlestop I guess, he called it VMax Stress throttling, which I'm completely unaware of. But I just checked TS thread and yep, it exists. It has something related to high voltage, I think you run the 10850K at high voltage for CML at 1.4-1.5v so probably that's why it's happening. I guess I think you should try it and fix that from Throttlestop. But EVGA should offer it in BIOS like ASUS does, currently I think they do not so that makes anyone to use 2 software products, Eleet and TS.

 

@Mr. Fox I use the PM2/PM3 slots only in Gen 3 mode. I run the GPU in Gen 4, I have never seen any whea errors on this platform. Very reliable platform, even after tearing up my cpu still running like a top.

Are you using NVME slot PM1?

 

It absolutely is the voltage issue. Because I can run a slower frequency with higher load line and it causes the same problem.

I thought I read about it some where coincidentally by randomly coming across it. And I could never find that person who had the problem again. So thanks for sharing.

 

I did not have that issue with downclocking above 5.3GHz.

Here is a BIOS mod with additional menus available for the Z490 Dark K|NGP|N mobo that I got from Lost_N_BIOS at win-raid forum.

 

That looks just like 11th Gen to me. Thick substrate and very thin die. My 11900K caps almost looked flush with the die on my 11900K. So I can certainly see how that big ceramic looking cap would ruin the show lol.

I run my 11900K without 16 SMD’s lol. The 12900K might be ok without 1.

 

One more rason to buy two boards and double up with ram sticks

 

If your on the bios 1.06 they might have fixed it.

 

This still cracks me up.....

 

@Talon JayZ2C's BIOS screen for his 12900k during power draw tests. SP79:

 

That modded BIOS is v1.09. Here is the thread relating to it. https://www.win-raid.com/t7539f54-Request-EVGA-Z-Dark-Bios-Unlocked.html

Even with the chiller, the 11900K gets too hot to run 5.3GHz in Cinebench. If I start out with the CPU at 6°C it still hits 90°C at 5.2GHz all core in CBR23.

https://hwbot.org/submission/4847913_

https://hwbot.org/submission/4847904_

https://hwbot.org/submission/4847900_

https://hwbot.org/submission/4847892_

 

What's the Vcore and pull under load with CB23 @ 5.3? Could just be a hungry bleeder.

 

The vcore is only like 1.475V and it will run something short, like wPrime 32M. But, it gets too hot for Cinebench or wPrime 1024M and has a watchdog BSOD when it overheats. There is just enough extra heat going from 5.2 to 5.3GHz to be an issue, but I think the delid will (hopefully) take care of that.

I am working on a Windows 7 clean install now. That is my top priority at this time, even more than a delid. It is imperative that I am not left with the cancer OS as my only option. That would suck beyond words.

 

I reported the WHEA Errors during testing. They reproduced them. I have no contact with that team so I do not know when there will be a fix.

 

ASUS released 0x50 microcode update for their boards for RKL Processors last week, also the latest beta BIOS which packs that on Maximus Apex XIII supports Win7 and XP. Apparently previously an old SuperPI exclusive one only had Win7/XP support. I think we do not have that new microcode on EVGA yet so a BIOS update might be coming soon.

 

I wonder how that chip looks hooked up to an oscilloscope with all the filtering caps gone BuildZoid would probably get a panic attack lol.

 

One thing I do like about Asus and they stand heads and tails above other motherboard makers is their activity in the forums on OC (and their own) and the rapid releases of public and customized BIOSes for their upper end motherboards.

 

Since manually setting all PCIe lanes to Gen 3 there has not been one single WHEA error. Later today I will change the slot for the GPU back to Gen 4 and see if the situation continues to hold true enabled for that slot. At this point in time there's really no clear advantage to having Gen 4. Maybe someday it'll be relevant but right now, not so much.

What is really strange is I have all of the correct hotfixes, updates and drivers for NVMe support that are normally required and Windows 7 cannot see any of them. It's like they don't exist. There's no trace of them in Device Manager and I can't access them. This is why my prior Windows 7 installations would not work on this new Z590 setup. Installed it on a 2.5 inch SATA SSD with ease.

Doing a clean install of Windows 7 was so much easier than it was on the AMD system. It's is functional now but I'm still on the hunt for some drivers. These are things that Windows 7 Intel chipset drivers would resolve. There's one item it causes a BSOD when I install the driver. I can't tell what it is by name but I have to use Windows System Restore in repair mode to recover from it.

I'm probably not going to burn too many calories on that because most of those "system devices" with problem code 43 don't actually matter in the grand scheme of things because they serve no useful purpose. I haven't looked for a compatible Intel ME driver because I generally disable that in the BIOS. Intel ME (and likewise DTPF) is a totally worthless telemetry and remote access piece of crap that has no reason to exist in a consumer environment in the first place. And there is certainly something positive to be said for preventing the operating system from having any ability to access firmware settings. The less influence and access the operating system has relating to firmware, the better off we are.

 

Well, 11900K can RIP. I followed the procedure exactly and the delid turned into a de-die. The die top stayed soldered to the IHS and lifted off exposing the CPU cores. Pushed the IHS as far as it would go with the delid tool and baked at 170°C (350°F for 25 minutes), lifted it up and this is what I was left with. I've never seen anything like it. I am going to guess that pushing it with the delid tool may have done it since the top surface of the die is so thin.

 

It looks like you lifted it off too soon. I pushed it back and forward in the delid tool until friction broke the camels back. Kinda like bending a coat hanger back and forward. Once I was certain the solder was weakened, I pried it off with a debit card.

I did bake in the oven, delid tool, bake in oven, rotate direction then delid tool.

But this is why I ripped my little caps off, because I kept going back to the delid tool and going a different direction to weaken that solder. After it finally gave way on me. Only, I got lost on which way I was going, and went a tap too far.


I’m very sorry for your loss. That direct die would have opened up 150-200Mhz easily. With huge temperature reduction. I hate that happen. I know the feeling. I thought I killed mine for sure.

 

Speaking of misfortunes, I cracked and destroyed my Clevo X170SM-G's screen today. Silly me used too much force to pull the screen out of the housing.

I was expecting the screen to be screwed into the housing like on my Alien, only to find it had been glued in. I wedged a card under the screen to try and pry the glue off before taking the screen out, but I think I used too much force as the screen was flexing quite a bit. Had I done this to a screen like my Alien's, it would not have broken. The screen on this laptop is too thin. I felt like I was being really careful too! Evidently, I wasn't careful enough.

For those of you wondering why I removed the screen in the first place, I was going to soak it in water using paper towels to soften up the glue between the screen and the anti-glare matte layer. This would allow me to easily peel the anti-glare matte layer off, thus turning the screen into a glossy screen, which is what all computer screens should be. I absolutely HATE matte panels and cannot deal with them. I tried dealing with this one, but I couldn't take it anymore. I had to turn it into a glossy display like my Alien's.

Welp, there goes the cost saving measure I was trying to employ. Now I have to buy a new screen, but at least I can get a glossy version of this screen directly from a vendor. I think gluing the screen into the housing is a stupid design. Clevo should've screwed it in instead, which makes for easier maintenance and a much lower risk of something like this happening.

 

Yeah, it really sucks. But, life goes on. I've still got plenty of things to be thankful for. Live and learn. What started me on the slippery slope was the poor judgment in selling what I had and moving to Ryzen. If I could undo all of it I would. I'd be about $2000 richer and less frustrated. BUT, I'm more dangerous than I was before because I know more than I did when I started. Kind of like using W11 gives me license to bash it. Same concept here, LOL.

I've already taken the 10900KF and put it in the Z590 Dark and bought a $350 10700K from BestBuy that I am going to to go pick up from the store to use in my work desktop. I'm not going to waste money on another 11900K.

I don't think I lifted it off too soon. I really do think pushing it with the delid tool caused it. I pushed it one direction, then rotated 90° and push it sideways. When I did that I hear a faint crunching sound that sounded a little bit like crushing a small piece of glass. I was instantly worried when I heard that. Perhaps there was a manufacturing defect and the die was not secured correctly. It is so thin it wouldn't take much to break it off exactly like what happened. You can actually see where it is shattered on the one longer side in the middle.

I think it's probably getting close to time to excuse myself from spending money on this dying hobby that was fun for so long. It's less fun and there is more defective garbage and nonsense to deal with. Even Windows 10/11 are getting close to enough to be a valid reason to abandon it because they are such horrible trash. I'm definitely sick of playing the lottery and I am not into the AI robot dumba$$ firmware automatic overclocking thing that is becoming popular.

 

You have my condolences as well, brother. This weekend must be cursed, LOL. Halloween was last weekend, so what's up with that? Maybe it's hell week.

 

Ok, I only rotated 180 degrees.

Hey, I’m willing to sale my 10850K for a great price if your interested includes die frame too. It is already delidded. It’s a 5.4 ready beast. If it’s not too late.

 

Thank you. I've already got it sorted. I'm at BestBuy right now picking up a 10700K to replace the 10900KF I moved from my work computer into the Z590 Dark. I was probably going to do that anyway because I didn't really care for 11900K. But I had intended to put that CPU into the Strix Z490 and it makes me sick what a monetary loss that was. Oh well... as they say, "You can't take it with you, and in the end it's all going to burn." I've taken quite the financial beating over the last month-and-a-half with trash tech and I'm ready be done with that.

 

I would have let you have it for $300 with your old direct die frame too. Let me know if you change your mind.

Yeah I wish you could have seen the 11900K shine once direct die though. It’s a stable processor, as is that motherboard. At least you’ve got a great motherboard.

CPU’s die, break, and things happen. That 10900K I got from you was great for like a week. And it just randomly died out on me lol. So it really does just happen randomly. Which was the only reason I got a 11900K.


I’m almost starting to think how well the factory solder is done plays half the song within winning the silicon lottery.