Coachdude

It's just where the pads of the cpu are making contact with the pins of the motherboard socket, nothing to worry about. Think of it as a "witness mark", more or less.

To add to what the others here have said, it looks like your motherboard isn't quite mounted properly, judging by the angle of the audio jacks on the bottom, they look askew to me, and that's probably what lead to the damage to that specific USB port in the first place. So if you feel comfortable remounting it, I'd definitely go ahead and do so...

Bro you've got such a good system there already, why waste money on anything right now? You paid good money for your system as is, get your money's worth out of it. I'm still using a 3900X which is already a good bit slower than your Zen 3 counterpart, and it's still getting the job done fine. As long as your PC does what you want it to, why worry about the new things? Don't get caught up in the FOMO, IMO.

I'm wondering if the latest bios agesa has something to do with it now. IIRC Zen 2 was vulnerable to a recent security vulnerability dubbed "Zenbleed", so I wonder if the latest bios potentially has the security fixes applied and is lowering performance. If that's the case there's not much to be done unfortunately. I've a Crosshair 8 Hero with the second to latest AGESA and I've not noticed any degradation in performance so far. So it could be that. I'm not sure what if anything else it could be at this time. Sorry I couldn't be of more help. If I get around to installing the latest bios and thus agesa on my board I'll let you know if I notice any performance drop off.

Those scores definitely look a bit low to me given the clock speeds to be honest. Would you mind listing the rest of your system specifications? Your motherboard and ram especially. It could be something as simple as a bloated Windows install. It could be an out of date bios, using an ancient AGESA version or something along those lines. Have you made sure to enable your RAM's XMP profile in bios? I'm not going to lie to you it's kind of difficult to assess these kinds of things through text, without having a hands on with the system in question I'm kind of just spitballing here. All that being said. Make sure your motherboard's bios is up to date. Make sure you've installed the latest chipset drivers from AMD's website. Check task manager and make sure there isn't anything hogging up the cpu. And make sure your RAM's XMP profile is enabled in bios. Beyond that I really can't think of why it'd be scoring so low. I mean it's not like those scores are bad necessarily, but they do look low to me given the clock speeds you're running at. Again it's difficult to assess these kinds of things through text. But hopefully one of these things points you in the right direction. I know I've had some wonkiness with some of my past systems before as well, and you really just have to go through the motions to try and figure everything out.

Alright, that seemed odd to me so I wanted to make sure. That being said, at 4.4, I would expect a bit higher in both R15 and R20. I would still test things at stock and see what you get. As the scores you posted from what I recall should be around stock levels. It's possible the 4.4GHz overclock isn't entirely stable, and thus is leading to lower scores potentially. Reset back to stock and test, and post back here what scores you get at stock, and go from there. Zen 2 wasn't really all that favorable to manual overclocking from what I remember, I have a 3900X myself and just enabled PBO as a set it and forget option. I would test what 8 cores on this chip does in R15 and R20 for you, but due to the CCX configuration I don't think it would be entirely comparable, as each CCX on this chip has only 3 cores spread across 4 CCXs. As opposed to two 4 core CCXs on your 3700X. So yeah, reset to stock, test. And then maybe try enabling just PBO and see what that gets you. I haven't much experience with manual overclocking so perhaps someone with more experience can guide you further in that area.

Those scores seem about right, but for Cinebench R15. Are you sure that isn't the version you're testing? R20 the 3700X should be around 5000 give or take a few hundred points. I'd double check which version you're running as 2000 seems to be about what a Zen 2 8 core scores in R15, more or less equivalent to a 9900K in that same benchmark... If you're actually getting 2000 in R20, it would seem to be you have a CCX disabled, and I would make sure game mode in Ryzen master is turned off, and reset bios settings to default. Make sure you see all cores enabled in task manager.

It's possible the two new ram sticks you bought aren't actually exactly the same, they could be totally different even if they're listed as being the same on whichever website you bought it from. I actually dealt with a similar issue a while back, using similar Corsair Vengeance 3200 stuff as well. Turns out when I went to upgrade from 16 gigs ( 2x8 ) to 32 ( 4x8 ), the new kit of ram I bought was actually a different revision from what I had previously, and whilst my system was able to post, I did have some memory stability problems. All this was solved by buying one certified ( 4x8 ) kit of G.Skill ram. I'm not saying this is what your problem is, but speaking from past experience it's definitely possible. Now, if you're saying none of the ram works and it did before, I'd tear everything down and double check everything. But if you're trying to go from 16 gigs to 32 and you're having problems, it could be what I described above. It's really hard to diagnose things like this without being hands on with the hardware.

12th gen is superior to Zen 3 bar the X3D variant. Between the 12700 and 5900x, the 12700kf will be faster in gaming and likewise VR, and even in multithreaded workloads. You really can't compare cache amounts between different architectures like that, with the X3D being the exception just because it has so much of it, and even then only in some titles some of the time. So again, if those are your two options the 12700KF will indeed be faster. All that being said, you may be able to find a better deal on something like a 13th gen i5 or the likes, which may end up being faster than even the 12700 given the additional clocks and L2 cache that 13th gen provided. Just make sure you're getting the most for your money is all.

Honestly, if you're wanting to stick with DDR4 and aren't planning on moving higher than a 6700XT, I'd just stick with the 9700K for now. Wait for the next releases from both AMD and Intel and see what your options are then. The 9700K is a bit dated sure, but it's still a decent performer if your main use is just gaming. Personally I probably wouldn't have gone with this particular i7 in the first place due to the lack of hyperthreading, but it does have 8 cores and can usually clock up to 5GHz+ or so, and with a modern mid-range card like the 6700XT, you should have a perfectly respectable experience in the vast majority of titles out there. I'd say hold off until you decide to make the leap to a DDR5 powered platform. Going from LGA 1151 to AM4 just doesn't make sense to me. It would be faster sure, but not an insane difference in my opinion. The X3D would be significantly faster in some titles yes, but you'd be going from one dead platform to another dead platform, and you'd probably be in the same boat in a few years down the line that you're in now. So again, unless you're willing to move to AM5 or DDR5 13th gen/14th gen, an upgrade for strictly gaming performance just isn't warranted given the money you'd spend to do so in my opinion. Just my two cents.

I would expect a bit higher than 3.6 in Cinebench honestly, but I do believe R23 and 2024 are a bit heavier than older versions, so perhaps that's just your cpus limit in those workloads. What kinds of temps are you getting? You can try enabling PBO in bios to see if the higher power limits allow it to boost further, but you're probably going to max out around ~4GHz or so with default PBO. As a 3900x owner, I can tell you I see around ~4.15 GHz in Cinebench 2024 with PBO enabled, so I wouldn't expect a 3600 to hit much higher than that without manual overclocking. 3000 series just didn't clock that high to be honest. But you are hitting at least 3.6GHz which is the 3600's base clock, so it is running within spec. So as far as I know everything seems to be working as it should. Definitely check your temps and try enabling PBO though, if you want a quick and easy way to get a bit of a boost that's where I'd start. I wouldn't bother overclocking with Zen 2 though to be quite frank. Any gains you may get would be negligible. Zen 2's biggest weakness was the split CCX design, and really no amount of realistic clock speed gain you'd get is going to offset that. That's where RAM and FLCK tuning comes into play, but that's a whole other thing entirely.

Which stress test are you using to test with? If you're running at the default power limits in something like Prime95 for example it will run into those limits and more than likely run at or close to base clock. 4.2 is the rated single core boost clock, and it will not hit that when you're maxing out all cores in a stress test like Prime.

Check task manager. Could be a program, driver, etc. There's just not much to go on with a monitoring graph alone. If you're playing a game or something that'd be normal but if you're just sitting on the desktop it could be something as simple as windows update doing something in the background... So check task manager and see what's using the most cpu time.

For what it's worth, I've been running a 4x8 kit of G.Skill 3600 with my Ryzen 9 3900x for about two years now, and I've yet to have any stability problems. XMP runs at around 1.35 for DRAM voltage, and soc voltage is around ~1.080. Granted this will obviously vary between cpus, motherboards, ram kits etc. This kit was on the QVL for my motherboard as well. As long as voltages are within safe limits, I doubt any noticeable degradation or instability will occur during the useful life of the system. "Silicon lottery" will always be a part of that though. That being said, Ryzen 3000 officially supports up to 3200 only, so it isn't a guarantee that higher speeds will work, but I'd say most 3000 and later cpus should handle at least 3600 with 1800 FCLK just fine. It was really only Zen 1 and Zen+ where obtaining higher speeds could be a bit of a chore iirc.

Yeah don't do that. Especially with an older gen Ryzen, either stick with what you've got, or try and get the same kit with the same revision and go all the way to 32 gigs in a 4x8 configuration. Try and make sure the revisions are the same between them if you can. I've had trouble in the past mixing even the same models of ram bought at different times, because ram is usually sold and guaranteed to run as a kit. But I've also had it work just fine in a different computer, so your mileage may vary... But definitely don't run your system without dual channel, the capacity increase isn't worth the performance hit that will incur with the config you're suggesting.

Zen 2 will definitely bottleneck a 4080 in certain scenarios some of the time, that being said, like always, if it does what you want it to do as in performing to your expectations fps wise and programs wise, I wouldn't mess with it. Choosing the right cpu will depend on rather you lean more towards gaming or more towards work related tasks, the 5800X3D will be the fastest gaming wise, but it'll lose ground in work tasks just based on the core count vs the other processors you've listed. I'd use the system as is with the 4080 and see how it performs. If it isn't up to snuff for you then consider your options and what you think will give you the most value for the tasks you mainly do. As far as overclocking goes, it won't make much of a difference with Zen 2. Zen 2 still had the split CCX design and no amount of clockspeed you'll realistically hit will make a difference worth noting. Zen 2 's weakness is the core to core latency as a result of the split CCX/cache design. Most of the time it isn't a problem but you can run into cases where the latency between cores just limits the maximum amount of performance/frame rate you'll be able to achieve.

Not really familiar with the game, but the 7700/K are old Skylake based quad cores, it isn't unheard of for them to be near maxed out in newer games. That being said it probably isn't a problem if you're satisfied with your current performance, and it's currently a decent pairing with the gpu you have.

The 7700K is getting a bit long in the tooth now tbh, but at 2K in newer games you'll probably still get a performance bump, though you'll be holding back the 4070 ti for sure with the old quad core. I'd probably look at 13th gen Intel if I was looking to upgrade personally. Zen 4 is an option too but I think Intel's current offerings are a bit better rounded especially in the mid range of the product stack like i5s to i7s. You'll probably still be able to squeak 60 fps out of most games with the 7700K, but multitasking won't lend well to it at all, and if you're running anything in the background 4 Skylake era cores just don't cut it any more. If you're going to upgrade the gpu, upgrade the platform as well, it'll be a much nicer experience.

The 3500X is a 6c/6t cpu, from what I've seen it'll be fairly typical for it to come close to maxing out in newer more multithreaded games. Same story for the 9th gen i5s. I'd say overall the 3500X with a 2060S is a decent pairing, I really wouldn't have any complaints about such a system myself. I run a 2060 reg and it runs everything I need it to just fine. If I were you, I'd probably scale back the Raytracing and just enjoy the game with a bit lowered settings, as it still looks good even with settings lowered in my opinion. If you find yourself needing a bit more cpu grunt in the future there's always Zen 3, but really I don't think it's warranted with a 2060S. Like I said there will be some games that are limited by the cpu some of the time, but there will also be plenty of others that aren't, and max out the gpu instead. I'd say if you're happy with the performance currently, don't worry about it, just enjoy the system for what it is.

That doesn't seem possible unless they're running faster ram than you are or something. Even then it shouldn't be that big of a difference, do you know the complete system specs of the system you were comparing to? 3200Mhz should be more than sufficient to get the majority of performance out of Ryzen 3000. I run 3600 myself and it tends to do well, but I had 16 gigs of 3200 before that and it was only marginally slower on average. So unless something is just configured weirdly in your bios or something I just don't see it making that big of a difference.

As long as your current X370 can run the R9 sufficiently, I wouldn't bother "upgrading" to an X570, from everything I've read it should be able to power a 3900x just fine. You'd be spending money for little reason on a now outdated socket, you'd be much better off in my opinion saving up a bit more and doing a complete rebuild if you're already going to be replacing the motherboard anyways... Only you can determine if that's worth the expense, but just replacing the motherboard just doesn't make any sense to me personally. This board has AGESA support for Zen 3 as well, so if you think you're limited by the cpu you can go that route as well. A 3900x should have little trouble maintaining good framerates in CS:GO, but I don't play that game so I can't really comment on it. I have a 3900x as well and I haven't found it limiting me in anything I personally do with the system, but like I said that's going to vary on a game by game and by program basis. Zen 3 made good gains due to the CCX redesign, but I would guess something else is being screwy before the motherboard if it's limiting performance to the degree of being "exceptionally bad."

I think the 5800X3D is probably going to be your most logical choice here, assuming you're wanting to keep the same motherboard and ram. If you do want to switch it out, I'd probably go 13th gen Intel personally, as from what I've seen they're just better all around processors than most of the 7000 series Ryzens, and you do have the option of going with one of the DDR4 based boards as well, but if you're going to be switching out the board I'd personally go all the way with DDR5, but that's just me. Regarding PCIE gen 3, I don't think it'd make much of a difference for a 3080, maybe a few percent here and there, but it wouldn't be drastic. In the end I'd probably just go with the X3D and be done with it. Assuming this is purely gaming as there will be bigger differences between 13th gen options vs the X3D if you're going to be doing work/production tasks with it as well, but that's up to you to decide.

It looks like you have an unstable overclock applied either in Ryzen Master or in the motherboard's bios. Reset both to default and just enable XMP, or DOCP as it's sometimes called. Should fix things. Also you really should consider getting a 16 gig kit for dual channel ram operation, you're really hampering your potential performance with a single 8 gig stick like that...

Happy to help friend. Take care, and have fun. =)

@superfantastic Higher resolutions tend to shift the load onto the gpu more yes, because it's literally having to render more pixels than vs. a lower resolution. Regarding graphics settings, turning them to Ultra will usually hit the gpu harder yes, but some can also burden the cpu more because they increase the number of draw calls and simulations the cpu has to work on, think crowd density in the Witcher 3 or Cyberpunk for example. It isn't always so cut and dry. But honestly I think you're overthinking things a bit. Your hardware is more than capable of pretty much maxing out a great majority of games that currently exist. I always set things a tick or two lower than Ultra just because the performance trade off is hardly worth it most of the time, and because my gpu isn't really all that powerful to begin with. But if I'm playing something like Battlefield 5 for instance, I don't really care that the grass looks more pretty on Ultra vs low, because that isn't going to help me acquire targets any easier. So I put that particular setting to low because it helps with visibility. It all depends on what you want out of the game. Is it a single player story focused game where you can afford to turn graphics up in exchange for framerate? Then by all means do so. Or is it a fast paced multiplayer shooter where being able to easily spot enemies at a distance will often times win you the engagement, then lower the settings necessary to achieve that. At the end of the day its all up to you, there's no right or wrong answer here. It's what makes the flexibility of PC gaming so great imo, you have all those options at your disposal. =) Keep in mind I hardly claim to have all the answers regarding graphical settings in regards to gpu vs cpu loads. Different game engines perform differently, and there's just so many out there that whatever I say will just end up being anecdotal to my personal experience regarding games I have played, and that could be totally different from the ones you play. Personally, I just lower and disable things I don't like, like motion blur or chromatic aberration, or until I hit my monitor's target refresh rate, which for me is 144hz, and call it a day. Most of the time games still look amazing at lower settings anyway, so it isn't really a big concern for me.