EightyDuce

Hahaha, well, not exactly. High end I meant those with BCLK/ECLK/AI OC unlocked. Which are the tippy top boards. Only B650 board with those features is the Gigabyte Aorus Master. Basically boards with auto tuning of voltages and EzOC features. Anyway, Derbauer just posted this video. It's by no means conclusive as he mentions himself, but the issue may involve EXPO profiles and specifically SOC voltage they may try to set automatically. Go in your bios and double check SOC voltage. Mine has been manually set to 1.25v from when I was tuning the memory. Still too early to tell but some things to check on in the meanwhile. TLDR no one really knows.

They rolled back 1410 back to 1409 dated 4/21. That being said, the issue "seems" to stem form automatic voltage being set by the overclocking module of the high-end boards so boards like the B650E-F are "safe". I still rolled my bios back to 1409 to be on the safe side.

Just FYI, unless something has changed since I played with RGB few years back, 3 pin is ARGB and is 5V, with 5V Signal and ground pins. 4 pin = RGB and is 12v with reg blue and green signal, plugging 5V into 12V is obviously no bueno. Long story short they are likely incompatible no matter how many pins you bend, you would need another solution, like new RGB.

I haven't had any issues on my B650E-F (currently on v.1410), but I may roll back to the 4/21 v1409 just to be safe.

It may not be as exciting as a graphics card or a CPU, but it's something that can hold that shiny hardware back if the airflow is terrible. There's also the layout, cable management and ease of maintenance (removable filters, especially if you just tuck the case away where it's sucking in dust). To be honest, it's hard to "visualize" the price difference, until you build in a good case then build in a bad case....then it jumps out at you. As for USB ports, nearly every modern case has a front USB breakout panel, some have traditional USBA, some USBC some combination of both. Biggest limiting factor for IO will be your motherboard. For just north of $100: Nzxt H7, Corsair 4000D, Fractal Pop, Kian Li 216. Under $100: Montech X3 Mesh...honestly, just spend more and get a good case, you thank yourself later.

Yeah, nothing definitive yet. Some have pointed to Asus' Ai OC profiles. Something to keep an eye on.

Your options for overclocking the 7800X3D are fairly limited as its max clocks are locked to 5050Mhz. If you are on a high-end Asus board that supports AI OC, you can squeeze 150-200Mhz more out of it. Lastly, where you could get the biggest increase in clocks is via ECLK (external clock generator) if your motherboard supports it (usually only high end boards). You can still tune PBO/CO2 to stay at boost longer and at lower Temps like you can with other Zen4 CPUs. If you want to learn more, check out this guy. He does some great technical videos on overclocking.

Kingston should be Hynix all aorund as they haven't used Samsung chips in as long as I can remember. XMP vs EXPO doesn't really matter, both will work in AM5 however BZs timings will be better than either. Sadly motherboard manufacturers are very conservative with their implementation of their "scaling algorithm" for timings. They just take the JDEC spec and schmear it on... Which results in very loose timings. I'm going to try this soon as I out the system back together. Replaced the 7800X3D that wouldn't post above 5800 on the RAM, new one seems to do much better. My 7700X was able to get a sizeable I crease in bandwith and drop in latency with BZs timings and then tuning them further. 1. Stock XMP/EXPO 2. BZ's timings applied 3. Tuned Daily prior to upgrade

I'd have to go back and look, but I believe the 5800X3D plazma benchmark posted back in the thread was done with the 180fps cap. I can go back and re-run my 7800X3D with the 180fps cap, however it's not going to be possible to make an accurate 1:1 comparison due to other hardware difference (GPU, RAM... CPU PBO/CO settings), even with similar hardware but different brand.

Not sure if MSFS2020 does something different with their DLSS implementation, but I find the glass cockpit and analog gauges quite easy to read.

That I do not know. The whole frametime/pacing/fps/1% lows and all that is a bit out of my depth.

Think I figured it out. Quick run at 1080 High Preset No Mirrors

Where do you set FPS to max 400?

If you have a track you'd like me to try at whatever settings, I can try do that. We just had out first kid so free time is a bit tight.

I don't see why that would be the case, other than trying to limit the troubleshooting areas. If it's stable there's no harm in it. That being said, the gains are negligible and to effort expended trying to get it stable above 6000 is arguably not worth it unless you're benching.

Popped the 7800X3D in place of 7700X and it wouldn't post above 5800. 7700X was testmem5 stable at 6400/2100 and game stable at 6400/2167....thats a little disappointing. Dialed timings in and got it close to where I was with the 7700X. I game on MT, CPU frametime went from 3.5-5ms to 1.5-3ms. Heavily GPU limited still, honestly can't tell a difference. Going to do some more testing, depending on how it performs in Msfs2020 I may go back to the 7700X and sell the 7800X3D.

I would start looking for possible issues with your system configuration. On a 7700X and a 4090, at my current settings (fairly high) I'm GPU limited all day long with 20%+ CPU headroom to spare on a Reverb G2. You could try up your graphics settings to see if you can push ahead to the point of being GPU limited, though it seems you're already at 99% utilization so not sure how much room there is. Anyway, you definitely don't need a 10Ghz (if one existed) CPU to not be CPU limited, especially in MT mode.

Not sure where the other person is getting information from, but I ran DDR5 on W10 for about 6 months before switching to W11 when I did a full rebuild. In those 6 months I didn't have any apparent issues. @72Stu could you elaborate on DDR5 and W10 vs. W11 compatibility or why you suggest running one over the other?

I'm running a G2 on AM5 without issues. Sample size of 1 though, so take that for what it is haha.

Interesting. So if I'm reading this correctly, and it's early in the morning and I'm on my first coffee, the 7800X3D numbers are accurate. However, 7950X3D are not and should, in theory, be slightly higher due to core parking issues and higher speed and v+cache amount? Unfortunately, I didn't see them actually present the numbers for the 7950X3D, just theorizing based on slightly higher clock speed and a bit more v-cache. I got my 7800X3D yesterday, sitting in my desk, but I won't have time to tinker with it until later this week due to having to drain the loop to pull the CPU.

Look at Gigabyte B650 Aorus Pro AX. This and the ASUS B650E-F were the two boards I was looking at, specifically for the IO. I ultimately went with B650E-F for some of the under the hood bios OC features and the fact that I'm more familiar with ASUS bios layout. But both boards have excellent expansion and rear IO.

Reviews are kinda slim, but here is a grab from Paul's Hardware for Msfs2020.

Just ordered mine. Should be here Monday.

Most reviews are up:

Looking at majority of the threads where folks aren't seeing significant improvements are those with current gen mid-tier or high-tier last gen GPUs and were already largely GPU bound. Being completely GPU bound now that CPU bottleneck is gone isn't going to change much in raw FPS. Those folks will benefit from DLSS far more than MT, however, MT still will have a significant impact in MP.