ValkyrieStar

Introduction
During a recent build using an SM951 SSD, I had some problems getting either version (AHCI or NVMe) to boot on a Q77 chipset board. Now obviously the NVMe version wasn't going to work, but I was curious to find out why the AHCI version would not boot. So here's my adventure to share with you all, and hopefully help you get yourself running with with new NVMe SSDs on an older system too!
 
Information
• This will ONLY work for boards which have UEFI! Old style BIOSes WILL NOT WORK!
• Do not try this unless you are willing to risk possibly bricking your UEFI, it is unlikely to happen, but I cannot claim responsibilty for your actions. You have been warned. If your board has "dual bios" feature, then you can feel safe knowing that you can experiment a little more.
• Links to original posts I found information on will be at the bottom of this post. Credit for this post goes to them, this is a consolidation of those posts / guides.
 
The Problem
• It is possible to connect an M.2 or PCIe SSD to an older PC with UEFI, but you will find that these wont be bootable. Windows installer will warn you that you cannot boot from the drive you may be attempting to install from, or a cloned OS will not appear in the boot menu. This is because there are a few modules that are missing in older UEFI's and manufacturers are not willing to update large ranges of motherboards just for a couple of SSD's (this could also be seen as planned obsolescence).
 
The Solution
• We simply add the modules ourselves!
• I have included a downloadable package containing the following;
  - NVMe modules required to boot from NVMe devices.
  - SAMSUNG_M2_DXE module required to boot from XP941 / SM951 (AHCI) devices.
  - AFUWINx64 tool to read/flash the UEFI.
  - MMTool tool to insert the modules into the UEFI.
  - UBU tool to allow updating other OpROMs in the UEFI.
• The package and working modded UEFIs can be found HERE (make sure your motherboard matches first).
• If you feel you might mess things up, feel free to send your unmodified Extracted.rom file.
 
Prerequisites
• Before you continue;
• Download the package HERE and extract it to somewhere easy to access (C: drive?). Package File Virus Report HERE.
• Update your UEFI to it's latest version and keep a fresh unmodified copy along with the files from the package (same folder as AFUWINx64 tool is preferable).
 
Guide
• Extracting and Modifying the UEFI.
 
• Updating other OpROMs (optional).
 
• Flashing the Modified UEFI.
 
Results
• You should now be fully able to select the drive as a boot device in the UEFI.
• If you are successful, please PM or post your original UEFI and the modified ROM, along with board model and revision (use CPU-Z to find that). I will add it to my download server, so others can use them as known working versions.
 
Credits
• Info about getting XP941 / SM951 (AHCI SSDs) bootable; http://www.win-raid.com/t1458f13-Guide-How-to-get-M-PCIe-connected-Samsung-AHCI-SSDs-bootable.html
• Info about getting NVMe support into UEFI UEFIes: http://www.win-raid.com/t871f16-Guide-How-to-get-full-NVMe-support-for-Intel-Chipset-systems-from-Series-up.html
• UBU Tool: http://www.win-raid.com/t154f16-Tool-Guide-News-quot-UEFI-BIOS-Updater-quot-UBU.html

Introduction
During a recent build using an SM951 SSD, I had some problems getting either version (AHCI or NVMe) to boot on a Q77 chipset board. Now obviously the NVMe version wasn't going to work, but I was curious to find out why the AHCI version would not boot. So here's my adventure to share with you all, and hopefully help you get yourself running with with new NVMe SSDs on an older system too!
 
Information
• This will ONLY work for boards which have UEFI! Old style BIOSes WILL NOT WORK!
• Do not try this unless you are willing to risk possibly bricking your UEFI, it is unlikely to happen, but I cannot claim responsibilty for your actions. You have been warned. If your board has "dual bios" feature, then you can feel safe knowing that you can experiment a little more.
• Links to original posts I found information on will be at the bottom of this post. Credit for this post goes to them, this is a consolidation of those posts / guides.
 
The Problem
• It is possible to connect an M.2 or PCIe SSD to an older PC with UEFI, but you will find that these wont be bootable. Windows installer will warn you that you cannot boot from the drive you may be attempting to install from, or a cloned OS will not appear in the boot menu. This is because there are a few modules that are missing in older UEFI's and manufacturers are not willing to update large ranges of motherboards just for a couple of SSD's (this could also be seen as planned obsolescence).
 
The Solution
• We simply add the modules ourselves!
• I have included a downloadable package containing the following;
  - NVMe modules required to boot from NVMe devices.
  - SAMSUNG_M2_DXE module required to boot from XP941 / SM951 (AHCI) devices.
  - AFUWINx64 tool to read/flash the UEFI.
  - MMTool tool to insert the modules into the UEFI.
  - UBU tool to allow updating other OpROMs in the UEFI.
• The package and working modded UEFIs can be found HERE (make sure your motherboard matches first).
• If you feel you might mess things up, feel free to send your unmodified Extracted.rom file.
 
Prerequisites
• Before you continue;
• Download the package HERE and extract it to somewhere easy to access (C: drive?). Package File Virus Report HERE.
• Update your UEFI to it's latest version and keep a fresh unmodified copy along with the files from the package (same folder as AFUWINx64 tool is preferable).
 
Guide
• Extracting and Modifying the UEFI.
 
• Updating other OpROMs (optional).
 
• Flashing the Modified UEFI.
 
Results
• You should now be fully able to select the drive as a boot device in the UEFI.
• If you are successful, please PM or post your original UEFI and the modified ROM, along with board model and revision (use CPU-Z to find that). I will add it to my download server, so others can use them as known working versions.
 
Credits
• Info about getting XP941 / SM951 (AHCI SSDs) bootable; http://www.win-raid.com/t1458f13-Guide-How-to-get-M-PCIe-connected-Samsung-AHCI-SSDs-bootable.html
• Info about getting NVMe support into UEFI UEFIes: http://www.win-raid.com/t871f16-Guide-How-to-get-full-NVMe-support-for-Intel-Chipset-systems-from-Series-up.html
• UBU Tool: http://www.win-raid.com/t154f16-Tool-Guide-News-quot-UEFI-BIOS-Updater-quot-UBU.html

A 3900X is basically two 3600X's on one package; 3+3 & 3+3
 
CPPC Prefered Cores works well for a single program which uses only a couple of threads, they get locked to the fastest cores on the fastest ccx, no latency from them hopping around. You'll get better performance from it having higher boost clocks on those cores.
 
The issue is that as soon as you get into things that use many more threads, or multi tasking with more than one thing running at a time, they all get bunched together and choke up on that single ccx. So for my work, it's best to keep it off. Might see lower single thread performance, but unless you're actually 100%ing a single thread, it's not going to matter.
 
The only time you'll see that is in synthetic benchmarks. My CPU lost a whole 2 points in Cinebench single thread by disabling it. consistent 206 score to now a 204 score. It'll make more of a difference on CPUs where some cores clock significantly higher than others (mine all hit 4.4 except for 2 which hit 4.375 - they will do 4.4 too but only after coaxing them out lol), but even from watching it hop around, it never hops to either of those two slowest cores, windows seems to know what it's doing for once!
 
It didn't affect multi-core score in cinebench at all, still hits 1660-1665, but now I consistently see better utilisation across all cores when doing encodes using ffmpeg etc and they are running faster too, surpassing speeds that the 9900KS managed whilst that was melting the VRM, and coming close to it when the VRM was under control.
 
Now.. what do i have that i can sell to get that ram bought before the offer ends...  oh yes.. an i9-9900KS and a motherboard that isn't safe to put it in lmao

I don't know, it probably helps when only doing 1 thing at a time, but as soon as you're doing more than 1 thing, those 2 or 3 fastest cores start capping out and stuff starts choking up and stuttering. Disabling CPPC stopped my Destiny 2 and CoD MW stuttering completely. With it enabled i'd also get stuttery youtube while running encode or simulations, but with CPPC disabled it's possible to watch youtube without stutters and the whole system feels more responsive in general. Wierd.
 
I've been eyeing up some 3600 @ 15-15-15-35 G.Skill ram, it's 50% more expensive than my current kit. But i guess it'd be a decent investment vs 3200 @ 16-18-18-36. DDR4 isn't going anywhere for a couple years at least, and once i've got everything in this system i won't be upgrading for some time except for GPUs. Plus i've a couple friends that could use my current RAM. Not sure though, opinions?

True lmao, disabling the CPPC feature really help spread the load on Destiny 2 and CoD MW, before they were completely choking up cores 0,1,2 and 3,4,5 were basically idle doing nothing, now all cores are averaging about 4.3ghz and im no longer getting any stutter from it choking up the CPU cores. GSync is doing a great job of smoothing any microstutter (if there is any) from it being across two CCX's.
 

 
Honestly, for me i have some 32GB of 3200 mhz, it'd be more hassle than it's worth (and more expense) to get more ram. I'll keep a lookout for any offers on some good low CL high mhz kits (maybe 3600 C15 or something). I'd just run them at XMP and call it a day.

Well I cured it of being scared of cores by disabling SMT, generally i run gsync so i cant tell if its improved anything, but stuff is being spread out across all cores now (its still barely tickling the last two cores unless i force it to though)
 
Because of how my encoding work is mostly 1080p, it doesn't scale much beyond 16 threads, and it performs much better with 16 cores vs 8core with smt, so my plan *was* to get the 4950X and disable SMT, but that has the wierd side effect of making sleep mode disappear. Wonder if that's something that'll ever get fixed.
 
I do know Ryzens SMT does scale better than Intels HT due to differing implementations and bigger L3 cache, but windows scheduler just seems to forget that SMT exists and uses both threads of the "faster" cores before using the "slower" cores. My 9900KS would spread stuff out fairly evenly between all the main cores and only after that would it start using the second thread of each core.
 
*edit while still writing*
 
I disabled CPPC Preferred Cores in the BIOS, now it's distributing stuff pretty evenly, it's even preferring to only load up one thread per core which is preferable too. I'll go as far as saying its actually running cooler and using less power (the whole cpu is sitting at about 3ghz in valorant, rather than 3 cores being pegged at 4.3-4.4ghz constantly), and i have all 12 threads again, nice! I guess i'll just be running two encodes at the same time to fully utilise the 4950X.. damn that thing gonna rip!
 
The preferred cores thing supposed to hint to windows to *prefer* using certain "faster" cores first, but it seems windows just dumps everything it can onto those cores and is scared to touch the other horribly slow crippled weaker cores that can only hit 4.375ghz instead of 4.400ghz 😂

I flat out ignore the motherboards "cpu temp" sensor, as it's often inaccurate unless it's reading one of those CPU sensors, in which case, i still ignore it anyway.
 
For Zen2 (Ryzen 3000);
 
CPU (Tctl/Tdie) = a wierd averaged thing, that the SMU ramps up with any spike of Tdie (even ones that aren't picked up by monitoring software), but ramps down much slowly, at idle this is often routine stuff like system logging or whatever, and can lead to annoying as hell fans which ramp up every few seconds, this is the one which produces a sawtooth like graph.
 
CPU CCDx (Tdie) = is the instantaneous hotspot temperature of die "x", on a single die chip there'll only be CCD1, on the 3900X or 3950X there'll be both CCD1 and CCD2, it's updated at something like every 1ms. Watching this can produce a very spikey graph at load loads/idle.
 
CPU Die (average) = a moving average temperature (not sure what kind of timescale it's averaged over, maybe 500ms or something) which produces a much more reasonable and readable graph and would also be a way better thing to base the fan speeds on rather than Tctl, but unfortunately that's not something we can set in the BIOS. It produces a graph similar to CPU Package on Intel.
 

 
As you see, CPU (Tctl) is producing a constant sawtooth graph, which makes for a really f*cking annoying fan ramping noise when just chilling watching youtube unless you manually tweak away from that ramping range. CPU CCD1 (Tdie) is the hotspot measured at that moment (since it updates so fast, HWinfo "misses" many of the spikes here). CPU Die (average) is an averaged reading of the Tdie. Ryzen Master actually uses the CPU Die (average) reading.
 
The other temperature sensors you mention are indeed, Intel or older AMD ones.

Yep, it sits nicely at about 4.1ghz consuming about 100w, actual temps are a little on the higher than expected side tbh, at around 72-75c ish. Full bore prime95 doesn't push it much over 85c though, at about 4.05ghz and 130w (1.27v), this particular chip seems to be quite power hungry from the looks of things, probably why it ended up with no X at the end of its name. I don't feel like i'll be able to get an all-core of more than 4.2ghz (i reckon that's as far as i'd get with about 1.25-1.275v being the max it'll take before being too hot), and in which case i'd be sacrificing the 4.4ghz lighter load speeds (i did a quick test of an overclock though, since there's no P-States, it just sits there ramming voltage into the chip) in favour for slightly better all core load speeds, despite not caring about this CPU a whole lot, i'd rather let it do it's own thing.
 
I spent a few hours yesterday tweaking memory, i managed to get it from the stock 16-18-18-36 @ 3200 to a nice 14-16-16-32 @ 3466 with no additional voltage, performance in benches was unaltered, and i was met with some wierd instability that memtest86, prime95 large ffts, hci memtest all failed to find, this instability wouldn't go away when dropping it to 14-16-16-32 @ 3200 or by increasing voltages, so i bit the bullet and ditched that and just went back to XMP rated timings, which by chance randomly started scoring better than they did before..
 
The only main gripe i have/had/ish with this board (and it seems with AMD boards in general) is that the POST times can send you to sleep. Default was something like 24 seconds, which was just absolutely painfully slow, memory fast boot does literally nothing (idk why, i tested cold boot, reboot, and warm boot and memory fast boot didn't affect it at all). I eventually got it down to around 16.5 seconds by disabling CSM, disabling the wifi, and disabling audio which i don't use (i have usb external card), so that was a nice improvement. I might email MSI see if they have any comment about why Memory Fast Boot isn't actually doing anything.
 
I figured i'd try setting the PCIe gen to gen3 for all the slots since i have no 4.0 devices, and that dropped post time to 10.6 seconds, very nice! Then i realised this also included the cpu-chipset link, and my two NVMe drives were being bottlenecked a little, i set the chipset up to gen4 and the post time didn't change at all, i then set all the settings back up to gen4 (rather than auto) and it maintained the nice 10.6s boot time, wierd. On a side note, don't enable the "pcie 4gb address mode crypto" rubbish, it increased post times by like 5 seconds and had some other non favourable side effects.
 
I'll have a go with tweaking some other bits and bobs at some point but i'm not sure what else there even is to tweak to improve post times. I'm just used to pressing the power button on my Z390/9900KS and about 4 seconds later i was at the login screen, post times in that were 2.7s, and windows fast boot bringing up the login screen almost immediately after that.
 
4950X gonna rip through some work really nice i bet...

Well, i did some more tweaking here and there, finally got the fan profile setup in a manner where it's not yo-yo-ing, i suppose, when i get round to custom looping the whole system later this year (once i get a nice gpu and 4950X for it), i'll be setting it based off water temperature vs ambient temperature, so it's a workaround for now.
 
For temp monitoring i've found that;
CPU CCDx (Tdie) is the spontaneous measurement of each die (im presuming it's hotspot temp),
CPU Die (average) reading is a smoothened out version of much the same,
CPU (Tctl) is just some mad yoyo wierd ass thing that some derp coded and they forgot to fix ever.
 
So for now i'm sticking with the CPU Die (average), as that's easiest to read, while it'd be nice if i could base the fan&pump speed off that, it'll have to do as it is.
 
While playing games and encoding (which sustains a better avg fps with less stutter than the intel system), the VRM hit an absolutely insanely meltingly high temperature of 57c (/s), pretty sure most of that was the GPU heating up the case, as now i've finished gaming and the encoding is still running (with a fairly warm room) it's now sitting happily at 45c VRM.
 
My encodes which usually finish with an average at or below 4.5fps, i've been hovering around 4.0 fps the past two encodes i've cared to check up on. Not a huge drop, and it'll certainly pick up a few once the 4950X gets dropped in  
With PBO disabled and AutoOC off it hovers around the 3.9-3.95GHz mark, PBO limits disabled about 4.0GHz, then AutoOC also set to +200MHz gets about 4.1GHz avarage.
With lighter threaded loads, all cores can achieve 4.2GHz with no AutoOC, and 4 of 6 cores hit 4.4GHz with AutoOC to +200MHz, the last two cores not far behind either. I must not have gotten too bad a sample to get that big of a boost with AutoOC. Single core perf on Cinebench R15 is almost on par with the 9900K/KS with 209, multithreaded isn't too far off either, with 1660.
 
 

Unfortunately the board and cpu still haven't arrived, it's looking like tuesday before i get them. Even if they do arrive monday i won't have time as my car is getting its new turbo fitted. Damn this coronavirus delaying crap  
 
The Zen2 IMC is good for quite high memory clocks, the infinity fabric starts breaking up around 1900 so to go higher than DDR4 3800 you must unlink the ram from the infinity fabric. There's two options when doing that, either run it at half to maintain low latency at the cost of less throughput, or run it at an odd ratio which incurrs additional latency but higher throughput.
 
The best option though is to stick at DDR4 3800 (or whatever the max is to maintain 1:1 ratio with infinity fabric), and get the timings as tight as possible, that way everything is running in sync for the lowest latency possible, which most tasks benefit from most. That'll also not limit core OC's in any meaningful way, so you'll achieve the most highly tuned, lowest latency setup possible.

Or my better solution of going to X570
 
Z490 was a serious contender, but the fact X570 goes up to 16 cores and will offer me unrestricted NVMe IO compared to it, that's what got me  

Or my better solution of going to X570
 
Z490 was a serious contender, but the fact X570 goes up to 16 cores and will offer me unrestricted NVMe IO compared to it, that's what got me  

Not sure what you mean here?
I've found generally MSI bioses have been well up to scratch. Budget MSI Gaming Edge AC has more BIOS features than the more higher end ASRock Phantom Gaming 9
Yep, i've no idea what anyone was thinking, Gigabyte was the only one who did decent VRMs on anything other than flagship but their BIOSes are just so awful i can't bear to use one (plus personally i hate their styling of their motherboards), MSI's Z390 ACE and Godlike were among the top, the budget ones weren't so great but still pulled above their weight. ASUS did some wierd stuff with a 4 phase Maximus Hero board which was far more expensive than any of their competitors equivalent boards. ASRocks whole Z390 series have bios issues with NVMe iops being really low (can be fixed without modding bios by disabling C-States but then the cpu runs hot at idle), and the VRMs are just outright bad.
Thanks! I'll update with pics once everything's arrived and in my system  it's irking me in the back of my mind that it's only a 3600 and not the 3950X and it'll be a slower than the 9900KS till the 4950X comes (unless i give in and get the 3950X lol), but i suppose at least the motherboard won't be trying to self combust 24/7.

Very much so, with AMD generally unless you're using sata based M.2 drives, NVMe M.2's dont take any sata ports away (unless they decide to implement all 12 sata ports and have 8 of them switchable with the two M.2 slots, which admittedly would be pretty cool!
 
With intel, depending on where they'ved mapped the ports to, You lose either 2 or 4 sata ports and you only have a max of 6 to begin with too, so you're left with 4 or 2 ports - not entirely handy. On intel using 3x M.2 NVMes leaves you with practically no chipset lanes left, maybe a couple of x1 slots if you're lucky.
If you want idk a 10gbit lan controller (pcie x4) then unless the x1 slots open ended (where it'd run slow), you're gonna have to use CPU lanes, limiting you to either 1 gpu, or 2 gpus but not being able to SLI/NVLink (crossfire would still work, but why are you using AMD gpus in an already expensive system?).
 
My ASRock Z390 board loses 4 sata ports when all 3 are using NVMes, which to counter this they added an asmedia chip for 2 more ports. Though I have to question their logic though. Thankfully not an issue for me though, as i only use NVMe drives, my case doesn't even have any slots for 2.5/3.5" drives.
 
The X570 Unify board has 3 M.2 slots which can all have NVMe drives simultaneously, you still have all 4 sata ports, and still have all other stuff active on the board. There are some X570 boards which juggle lanes about a little, but personally i'm not too sure why, even with 3 M.2 slots there's still 8 pcie (4.0) lanes floating around. Opposed to like 3 on Intel unless they sacrifice 4 sata ports then it's 5.
 
Some manufacturers are smart about how they manage their board resources (MSI X570 Unify/ACE), some seem to half try, then screw up, and others just were bad from the begining, my ASRock board is a proper mess with 4 devices all piggybacked off port #7.

It's based off the same design as the IO die, with some slight adjustments.
 
The 4 lanes are now 4 fixed sata ports, two groups of 4 lanes have the option of being used for another 4 sata ports each, or as pcie lanes, and the remaining 8 lanes can do whatever want with (lan, wifi, x1 slots, an x4 slot, or whatever). I'm not sure where audio comes into things though, i'd have thought it'd be built into either cpu or chipset, but i don't know.
 

 
Intel Chipsets have 16 lanes on the CPU, and up to 24 lanes on the chipset (depends on which chipset ofc).
Technically with Z390, if you didn't want any sata or lan, and were happy with only 6 USB ports (plus the 14 2.0 ports), you could have all 24 pcie lanes (the CNVi lan interface takes up one of those LAN lanes, not sure which, i guess it's possible to map it around somewhat like the normal lan ones are), the audio is on its own bus, unnaffected by the PCIe lane layout.
Generally though mobo manufacturers will put the M.2 slots on the lanes which can be mapped to Intel RST for hardware NVMe raid, my MSI board which only had 2 M.2 slots had the bottom x4 slot mappable to RST if you put an NVMe drive in there via an adapter.
One thing that does my head in though, is that the ports are enumerated in a wierd order, #17 first (M2_3), then #21 (M2_2), then #9 (M2_1).
Means i have to install my SSDs in reverse order for them to appear as disk 0,1,2 for drives C/D/E (i'm ocd like that lol).
 

Well, it's actually the first M.2 slot, any other M.2 slot(s) are from the chipset
 
 
Any AM4 board has the first M.2 slot direct from CPU (Ryzen has 4 extra CPU lanes vs Intels)
 
Any other M.2 slot comes from the Chipset.
 
The X570 chipset has 16 pcie 4.0 lanes
The B550 chipset has 4 pcie 3.0 lanes + 8 pcie 2.0 lanes
The X470 chipset and below, they have up to 8 pcie 2.0 lanes (X boards have 8, B boards have 6, A boards have 4).

Well DMI 4 (or PCIe 4.0 x4 as in Ryzen 3000/X570), is enough for two 3.0 x4 drives without any kind of limitations. The third M.2 on an X570 board is direct from the CPU, but on Z490, it's all from the chipset.
 
As for the IMC, i have 32gb of 3200mhz C16 ram, it's tried and tested on the 1950X (worked flawlessly but only after manually entering every single damn fricken timing - enabling xmp soft-bricked the board till i used bios flashback lmao).
 
I'd not be getting any more ram for the time being, maybe 64GB if prices are suitable, but probably not.
 
 
Or the fact they just completely skimping out on VRMs on Z390, my experience with Gigabyte bioses in the past has scared me away for the time being (X399 Gaming 7 was an absolute steaming pile of poop of bios problems). Had other manufacturers built proper vrms, I wouldn't be having the issues i am having now, and I would probably be completely happy with my 9900KS system.
 
With the X570 Unify, the entire "shroud" is actually part of the heatsink, it reduces VRM temperatures a further 5c vs the X570 ACE which this board has the same VRMs as, it's also a good £50 cheaper for me too, since i won't be using any RGB and i actually prefer the design of the Unify. The VRMs are also more than adequate even for the 3950X so i'm sure it'll be perfect for the 4950X too, unless the X670 Unify (if such a board comes to exist) tickles my fancy at that point too.
 
I probably could have waited for intel 10th gen, but considering what extent i had to go to to keep the 9900KS cool at stock with an undervolt, the 10900K can't be much better. That and the 3950X(and future 4950X) have a lot more cores, which i will benefit significantly from.
 
I suppose, if somehow, the sh1t hits the fan, i'll have a month to return the X570 Unify and 3600 for the Z490 Unify and an accompanying CPU - though it'd probably just be a pentium knowing intels pricing.
 
On a side note, this turned into a complaint about VRMs, into a "today i joined the red team", due to arrive this friday! I'm quite busy with work at the moment, and with my board constantly trying to self combust, i just can't risk downtime if it dies randomly.
 

For the most part i've managed to spend minimal amounts on upgrades by selling the old kit. The 1950X was a significant investment though, and also a very frustrating and dissapointing one. The arrangement of the cores meant my asetek cooler wasn't up for it, the enermax liqtech was brilliant, for 3 weeks till it broke, the replacement lasted a further 3 weeks till that too, broke. I looked around and it turns out they were designed poorly and corroded and blocked up quickly. Ended up being unable to run my 4.0/4.2 all core OC because the Asetek couldn't handle it, even stock was pushing it, that's how much the die layout mattered on those things. Stock clocks were buggy, often 1 or 2 cores would get stuck at max turbo and the rest stuck at 2.2ghz idle speeds, gaming was not the best due to the NUMA design to begin with, but it was far worse with clocks being bugged.
 
Call that 3 haha, DMI 4 though isn't out yet (that'll be 11th gen can hope), and even when it does arrive, it'll still limit with the third drive, to be completely unlimited, i'd need at least one of the nvme drives on cpu lanes, which is what the X570 offers, without the limitation of it taking up one of the gpu slots (when 30 series comes out i'm aiming for dual-gpu, which isn't possible if an ssd is taking up the second slot).
 
My only main worry is that X670 is probably due to arrive later this year. I suppose, if it does, and there's a better looking and featured board than the X570 Unify (i much prefer a clean monotone build than rainbow rgb unicorn puke builds), i could always just get that and it'd drop right in. If the X570 and 3600 turn out not great, i'd at least have some time to return them for Z490 if need be. Like when i went from Threadripper to 9900K, i'd rather have a slightly slower but rock solid stable system, than a buggy system which can be really fast when the bios decides to work but can also be absolute trash at other times.

For the most part i've managed to spend minimal amounts on upgrades by selling the old kit. The 1950X was a significant investment though, and also a very frustrating and dissapointing one. The arrangement of the cores meant my asetek cooler wasn't up for it, the enermax liqtech was brilliant, for 3 weeks till it broke, the replacement lasted a further 3 weeks till that too, broke. I looked around and it turns out they were designed poorly and corroded and blocked up quickly. Ended up being unable to run my 4.0/4.2 all core OC because the Asetek couldn't handle it, even stock was pushing it, that's how much the die layout mattered on those things. Stock clocks were buggy, often 1 or 2 cores would get stuck at max turbo and the rest stuck at 2.2ghz idle speeds, gaming was not the best due to the NUMA design to begin with, but it was far worse with clocks being bugged.
 
Call that 3 haha, DMI 4 though isn't out yet (that'll be 11th gen can hope), and even when it does arrive, it'll still limit with the third drive, to be completely unlimited, i'd need at least one of the nvme drives on cpu lanes, which is what the X570 offers, without the limitation of it taking up one of the gpu slots (when 30 series comes out i'm aiming for dual-gpu, which isn't possible if an ssd is taking up the second slot).
 
My only main worry is that X670 is probably due to arrive later this year. I suppose, if it does, and there's a better looking and featured board than the X570 Unify (i much prefer a clean monotone build than rainbow rgb unicorn puke builds), i could always just get that and it'd drop right in. If the X570 and 3600 turn out not great, i'd at least have some time to return them for Z490 if need be. Like when i went from Threadripper to 9900K, i'd rather have a slightly slower but rock solid stable system, than a buggy system which can be really fast when the bios decides to work but can also be absolute trash at other times.

My 360mm AIO is installed in the front, with 6x NF-F12PWM, at 80c on the CPU, the fans are all spinning away at 100%, there's three additional 140mm fans as exhaust, 1 on the back, two on the top, all are those ML140 fans (no LEDs) and are controlled by CPU temp too (ramping up to 100% at 80c), i recently cleaned out the AIO so that's all nice and clean. When the whole system is idle for some time, the CPU idles around 33c, vrm 50-60 ish.
 
I've already checked the heatsinks (i removed them and inspected the pads to ensure they contacted the powerstages - they do and quite well too) when i removed the plastic shield (which did help by making it take longer to reach the same temps).
 
For the time being i turned ffmpeg down to only use 8 threads, its slowed the encoding down by a pretty big lump, but it's now sat at 110-111c where it's constantly bouncing between 4.0-5.0ghz and making voltages go wierd, which seems to have an additional side effect of making doing anything feel incredibly sluggish for some reason. The burning smell has gone for now, but putting my nose near the back fan it still smells kinda not great.
 
I've been eyeing up the MSI X570 Unify for some time now, while initially i can't afford anything more than a R5 3600, it's got just about the best built VRM there is beside the ultra high end ones, more than adequate for a 3950X, so i may go down that route, RMA and sell off my 9900KS and replacement motherboard, and get saving for a 4950X when that arrives one day.

Jeez, someone woke up on the wrong side of the bed this morning..
 
As someone who jumped on board the Ryzen hype train and got a 1950X and X399 board, only to be plagued with a tonne of BIOS/AGESA related issues that simply were not fixed (still not to this day), I am still skeptical about Ryzen, it's far far from a polished platform, the number of bios updates that improve performance and fix bugs just shows that.
 
You all go "buy a ryzen and each BIOS update gives you free performance" yeah sure it does, but in reality it's fixing bugs which simply should NOT be there from the start. Go on all you like about how Intel has some security flaws or whatever, and true, i've lost 5-10% IOPs performance on my NVMe drives (when specifically benchmarking to test that anyway, otherwise no noticable difference), but there was no discernible difference in cpu benchmark scores that weren't inside the margin of error of said benchmarks.
 
My personal reasoning for going back to Intel from AMD was the fact that whatever motherboard you buy, whatever 9900K you buy, they will all boost the same, and perform the same, whichever damn BIOS version i use. Not only that, but I have 3 NVMe drives, and no Ryzen board could accomodate this (at least not with all slots at with the full 4 lanes of PCIe 3.0) at the time of purchase.
 
As for OPs question;
 
MSI MEG Z390 ACE
-Second best VRM
-Incredibly Solid BIOS, i've owned many MSI boards, and zero of them have had any kind of BIOS issue.
 
AORUS Z390 MASTER
-Best VRM of the 3 boards i list here
-Shittiest BIOS known to man, seriously, gigabyte somehow made their already kinda poor BIOS interface, even worse.. (idk how)
 
ASRock Z390 Phantom Gaming 9
-Weakest VRM of the 3, worthwhile doing a -ve vCore offset.
-Decent BIOS, a little buggy (i'm currently working with ASRock support to test and fix some of the issues), but very easy to naviage and setup.
 
I won't suggest the ASUS Maximus XI Hero, while it has a decent BIOS, it's VRM is weaker than some of the more budget boards from the likes of MSI and Gigabyte... and it's marginally more expensive.

Jeez, someone woke up on the wrong side of the bed this morning..
 
As someone who jumped on board the Ryzen hype train and got a 1950X and X399 board, only to be plagued with a tonne of BIOS/AGESA related issues that simply were not fixed (still not to this day), I am still skeptical about Ryzen, it's far far from a polished platform, the number of bios updates that improve performance and fix bugs just shows that.
 
You all go "buy a ryzen and each BIOS update gives you free performance" yeah sure it does, but in reality it's fixing bugs which simply should NOT be there from the start. Go on all you like about how Intel has some security flaws or whatever, and true, i've lost 5-10% IOPs performance on my NVMe drives (when specifically benchmarking to test that anyway, otherwise no noticable difference), but there was no discernible difference in cpu benchmark scores that weren't inside the margin of error of said benchmarks.
 
My personal reasoning for going back to Intel from AMD was the fact that whatever motherboard you buy, whatever 9900K you buy, they will all boost the same, and perform the same, whichever damn BIOS version i use. Not only that, but I have 3 NVMe drives, and no Ryzen board could accomodate this (at least not with all slots at with the full 4 lanes of PCIe 3.0) at the time of purchase.
 
As for OPs question;
 
MSI MEG Z390 ACE
-Second best VRM
-Incredibly Solid BIOS, i've owned many MSI boards, and zero of them have had any kind of BIOS issue.
 
AORUS Z390 MASTER
-Best VRM of the 3 boards i list here
-Shittiest BIOS known to man, seriously, gigabyte somehow made their already kinda poor BIOS interface, even worse.. (idk how)
 
ASRock Z390 Phantom Gaming 9
-Weakest VRM of the 3, worthwhile doing a -ve vCore offset.
-Decent BIOS, a little buggy (i'm currently working with ASRock support to test and fix some of the issues), but very easy to naviage and setup.
 
I won't suggest the ASUS Maximus XI Hero, while it has a decent BIOS, it's VRM is weaker than some of the more budget boards from the likes of MSI and Gigabyte... and it's marginally more expensive.

At least try updating the BIOS for that board.
 
Also make sure the underside of the CPU is nice and clean. Don't touch the pins in the socket though.
 
I've had a CPU that i bought as faulty because a memory channel was dead, turns out there was just some dirt on the pads which stopped the pins contacting properly, which happened to be something to do with the memory.

yes a bit better dac and better caps. Bass response is a little better. especially at high volume. That said even the Fulla 2 is awesome at 99quid 
 
oh and the 3 CoMmEs In BlAcK bEcAuSe ThAtS GaMiNg My BoIiIiI! 

It also depends on the headphones somewhat too.
 
I have a pair of Sennheiser HD 419's, closed over-ears. They are a little weaker on the bass side, but they are incredibly comfortable. Due to the nature of my work sometimes i can be sat working for almost 13 hours a day, they never hurt my head or my ears and my ears don't get too warm either. The weaker low/bass area makes them feel a little harsh but nothing bad by any standard.
 
I tried a pair of Beyerdynamic DT 770's, IIRC these were the 80ohm version, they were more pronounced in the low-end than the HD 419's and sounded much more flat overall, no particular spikes or anything. These were a nicer listening experience but they pressed on the sides of my head a little much for prolonged use, much more than 1-2 hours and the would give me a slight headache and hot ears and i don't have a wide head at all. Maybe could have been remedied by stretching the headband but they werent mine so i didn't want to do anythign to them.
 
I generally use a couple different songs to test different sound card and headphone combinations, to name just two;
 
This track i feel lets you experience the wideness of the soundstage.
 
And this lets you test out the bass reproduction. You easily tell if it's too weak or overpowered from listening to that song.
 
Both of those are in my favourites playlist too