Hempstead

A 9800X3D is selling on Amazon for USD $449, i.e. $74 dollars more than a 7800X3D. At that price diff, why would you go for the last gen Zen 4 as opposed to Zen 5 stuff? For a €3128.20 machine... adding USD$74 shouldn't be a problem, right?

Ah... I see it. Thanks! So, it's www.digitalcombatsimulator.com/Downloads need to have the file updated.

I am not seeing the document updated. It's still "DCS: F-16 Early Access Guide" dated Dec. 29, 2024 on the DCS web site. Can't seem to find the updated doc with sniper pod.

I have a static IP# assigned to my GamingPC2. And, my egress point IP# is also a static IP# (I pay quite a bit money for it monthly). So, no, there is no IP# change. And this has happened quite a bit before, like when I was in the mission editor, or browsing through my custom mission trying to decide which mission to play, or in Setting pages configuring my settings. It just kicked me out, claiming login expired. The only thing I can think of is DNS server glitch. Say, it tries to contact digitalcombatsimulator.com/.... to validate, but the DNS name could not be resolved for whatever reason... But that is very unlikely too, because once I got kicked out, login in a few seconds works fine means that is not likely to be the cause.

dcs.log.oldThat was the wrong log file... This is the correct one that it kicked me out. 2025-07-03 19:00:30.957 ERROR ASYNCNET (34508): The session has expired (401). Exiting...

dcs.log

It has been giving that "Login Session Expired" and kicking me out recently ever since the new launcher was deployed. Sure, restart will get me a login again. But... Today, it kicked me out when I was actively bombing some T-72s. Hey! WTF!!! I just got in less than 10 minutes ago! That's not a nice end-user experience! Shouldn't it renew my session authentication when I got into the launcher thing? Apparently not! How do you like to get kicked out while in a multi-player mano a mano dog fights? I am sure the other guy will not accept "the excuse" of getting kicked out! That's a potentially ego-dousing bug! Please fix it.

Minimalist frame is 2x 6' long and 2x 12 3/4" long 15x15 extrusion, plus 4 corner brackets and 8x screws. That's what I did. That's it. The other way is 2x 6' long 15x15, 1x 16" long, and 1x 4' long 15x15. This way, the two shorter extrusions will go under the 2x long extrusion instead on the same plane. And instead of using 4x corner bracket as you saw in the picture, you'd have to use 8 L-brackets to connect them. This way... the 4' long extrusion extends to the left and right, and acts as a stabilizing bar, and for mounting more stuff. Or, you can do 2x 6', 2x 4'. Or, 4x 6', 2x 4', (the middle 2x long ones are still the same thing as above.). There are unlimited number of variations... longer or shorter, different configurations, etc. but the central idea is long rectangular frame connecting the chair, stick, and pedals, and everything else connect to this central frame. You do you... And BTW, no vid for tipping over. I wasn't expecting that. Too bad... it was actually very funny.

The idea is already in the public. It cannot be patented, nor copyrighted... unless somebody else already copyrighted/patented it. Won't be me though.

I would caution anybody who build this frame one thing though. Width of the frame is necessarily the same center-to-center width of the chair rails. So, it's very narrow. Your tilting around to look at your six in VR is going to cause this thing to tip over. So, you would have to add a lateral stabilizing bar to prevent tipping over. Ask me how I know that! My solution to this is adding a cheap 4x4 deck post under the frame extending to left and right of the seat (right under the seat), yet another weekend trip to HomeDepot. And then added some office chair caster wheels (front stabilizing bar can be shorter). This kills two birds with one stone. Now this thing moves if I need it be... say vacuuming?

In the picture, that's what it looks like... mount the chair rails on the two long extrusion.... then, cut the front and back connecting extrusions, screw on... That's it. IMG_0363.jp2

I currently have two 2'x4' 3/4" plywood stacked and screwed together, topped with a thin aluminum sheet for smooth feet movement. And then the whole thing is screwed down to the 80/20 extrusion frame of my "pit" with some cheap woodworking tracks/screws. Actually, this was the original idea of my "pit", just a rectangular frame with a chair and a pair of pedals bolted on. I just added wheels, side rails for mounting panels, and front "gates" for mounting front panel and stuff. (I constructed the 3D models of the frame after the fact. It's an I-got-a-rough-idea-so-just-build-the-damned-thing design. ) PJ. It looks like you have a "chair" sitting on the carpet. Why not buy some cheap chinese-made 80/20 clone aluminum extrusion from Amazon, and make a simple 4-pieces long rectangular frame to bolt your chair and boards on? This way, you can use a chair that can slide back and forth (I do, and it makes getting in and out of the pit so much easier than a fixed chair), and so can the pedals and stick be slid back and forth independently. You can stop right there, or you can keep adding stuff like I do, still do. The frame should be made of at least Metric 40x40, or Imperial 15x15 extrusion, just in case one day you want to lift it off the floor, and it's strong enough to support your weight, and doesn't twist like crazy as Imperial 10x10 extrusion will do. The frame can be assembled in 1/2 hr if you have the right tools and parts.

I have been building that for a while already, except I write my own USB firmware for them, with home build accurate panels (mostly 3D printing and laser engraving), instead of tablets. Actually, quite a few people have been doing the same thing. For this scheme to work well. You need two things. 1. a way to see the real world and the VR world. For instance switch to camera view and back to VR 24 time a second, or build a transparent overlap with an OpenXR layer. Otherwise alignment is a bit tricky. This is made even more challenging by DCS’s recent change to one shot recentering of the VR, instead of continuing recentering while you hold down the recenter button until you release the button. 2. You need a way to see your finger tip in order to do the feedback loop to navigating your finger tip, real or representation in VR, accurately. Quest’s hand tracking is not there. A few cm off, and refresh rate is too low, resolution is too crappy.

Oversized good power supply is not a bad thing to have to begin with. If money is not a constraint. And... My previous gaming rig's 1,200W power supply is from the previous previous gaming PC. Or, was it from the previous, previous, previous gaming PC? Anyway, it predates VR and had dual graphic cards, hence the oversized power supply. You know, from the heyday of SLI and stuff. It would have been my current gaming PC's power supply but I decided that I can use the previous gaming PC's dual boot Linux for compiling stuff, and its Win10's current DCS setup for AH64 gunner "passenger" ride "demos" with a cheap Quest 3S. So, I bought another way oversized expensive power supply, 1,600W. And note that... I don't turn off my computers. So, that darn thing has been running 24/7 for many years. > 10 years? Maybe the "low load" has something to do with the longevity? Don't really know, but I suspect it's a factor. And of course, key is a good durable power supply. You don't want those that will have capacitors blown a few years down the road. This... is difficult... by reputation, experience, and luck only. YMMV greatly.

Quest Controllers work. I recently build a new 9950X3D machine, about two weeks ago, and reinstalled everything from scratch. My Quest Pro's controllers still work. IMHO, Quest controllers are far superior than mouse. Of course, it's a very personal thing. So, try it out for yourself first and don't let others, me included, sway your decision without trying it.

Not that much... but because it's there, and it can be done. That's how progress is made. One small step forward at a time. It used to be 4x DDR5-6000 was dicey too. Now.... if it can be easily done, and the price difference is insignificant... why not?

Ya... Power over Ethernet. It's a standard that is build into PoE capable switches to carry power over certain wires in the Ethernet cable. The receiving ends have to split the power out. This way, you can skip the power wire to your Ethernet-to-USB-C adapter. So, one Ethernet cable in, and that's it. Obviously, you need an PoE capable Ethernet to USB-C adapter. I have one capable of the required 60W. Note that, PoE capable switches are not cheap.

Has anybody tried PoE?

Note that I was fully prepared for multi-week tuning sessions, but all I did was updating the firmware to the latest, and changing the Memory Bus Speed from Auto to 3,200MHz, as Auto resulted in 1,600MHz, so I forced it. Of course, after Win11 Pro installation, update to the latest drivers from Asus Driver Hub (to be turned off, as there is an RCE reported, and supposedly fixed), Windows Update to the latest, and Nvidia App/driver to the latest as well, the usual. That is Joe-Six-Pack achievable. It passed an one hour CPU + RAM stress test... There is not much point in running a 48-hr burn test I used to do on gaming rigs. They are not file servers. And I have had it up for about two+ weeks with multiple DCS/MSFS 2024 sessions, without a problem. I don't turn off my machines. It hasn't crashed once since. Also, before updating the firmware to the latest, it was very unstable. So, that tells you that the firmware improved quite a bit during the several months Asus made the board, and it being shipped to my hands. The "problem" at least seems fixable with firmware updates (plus the DIMMs improved; there is a good reason why I paid premium for Crucial/Micron DIMMs; they are more conservative and rigorous in what they claim, unlike some Asian startups.). So, whatever you "heard" on the Internet about 4x DIMMs being unstable is probably outdated, at least for this particular motherboard. And since this firmware is an AMI... I am speculating that all the AMI firmware equipped motherboards would be fine. Just speculating.... unconfirmed. The salient point I am telling you is that there are reports that 4x DDR5-6400 DIMMs are workable now, and I have replicated the result, easy. So, now you get an expanded baseline with 4x DDR5-6400 DIMMs. There is, of course, always some risks involved building your own PC, and be the manufacturer of your own PC. What's the failure state of taking a well known good baseline list, change one thing and one thing only, the 4x DDR5-6400 DIMMs in this case, and push? The worst case scenario is that the 4x DIMMs are unstable, no matter what frequency I set it to, and it corrupts my SSD. All I have to do is taking out 2x DIMMs, and perhaps tuning down the memory speed. All I lose is a bunch of time, and perhaps $200 worth of RAM, which I could either resell and recover some money, or use them in another computer. Risk, sure, not that much, really. All good clean fun, anyway. If you are building your own PC, and you don't have 40 years of PC building experience under your belt, then take a confirmed well known good list... and build just that. If you are more adventurous, take a baseline list, change one thing and one thing only and push. Nobody says you can only push once. You are welcome to push n times. I'd like to know your n-pushes and results though. Me? I am here for the games... I pushed one time... and I won... I wanna play some more games.

Watch this.

Just passed OCCT CPU + RAM stress test for one hour with flying color (and I have been flying DCS and MSFS 2024 with it for about 2 weeks). The CPU Tdie temperature never went over 60ºC, staying at around 58ºC (core temps were at some cool 32ºC). RAM temp was at about 42ºC. That's 32x logical cores going full blast at 100% for an hour. System Spec: Motherboard: Asus ROG Crosshair X870E Hero CPU: AMD Ryzen 9950X3D RAM: 4X Crucial DDR5-6400 DIMMs, 128GB (not on the MB QVL, but Crucial site confirmed the compatibility) SSD: Crucial T705 4TB Gfx Card: PNY RTX5080 OC AIO Cooler: Asus Ryujin III 360 Power Supply: Seasonic Prime PX-1600 All fans are Noctua NF-A14x25 gen 2, the AIO has 3x Noctua NF-A14x25 gen 1 (very quiet) Note that, the T705 SSD is a PCIe gen 5, and so is the gfx card running with PCIe gen 5. The T705 does run hot... 50ºC at idle. I am thinking about adding a fan just for it. But a gaming rig's SSD workload is not very high. It's very bursty but not very much sustained load. I would agree that the X870E had a very rocky start with PCIe 5, particularly with RTX5000 cards when it just came out. And it also had some trouble with DDR5-6000+ RAM and 4x DIMM at the beginning. When I first got the system up and running, it was very unstable... but as soon as I updated the motherboard with the latest firmware from Asus. It was rock solid, except Win11 and Nvidia driver still has that problem of once the screen blanks it won't come back, including the installer of Nvidia driver Nvidia App itself. I RDP into it, and it won't come out of the blank screen when I logged off RDP. But other that, it's been rock solid. JayzTwoCents was just testing DDR5-7800 (unstable) and -7600 (stable) for his 9950X3D rig today. -6400? Right... Of course, YMMV. And.... BTW... I didn't change a thing in BIOS. Didn't bother tuning anything.... yet. You have to establish a baseline before you push the envelop.

Keyboard and mouse, where to put in my pit? https://blog.hempstick.org/2025/06/where-to-shove-my-keyboard-and-mouse.html

I have 4 sticks of DDR5-6400, 128GB, on an Asus X870E Hero, Win11 Pro. It was very unstable... until I burned the newest firmware on the motherboard. It's..... kind of stable now. Not sure if it helps DCS at all. I have no baseline to compare it with other than I can get 45fps min. < 1,000' agl, and 90fps > 1,000' agl in Germany map, or any map, with a Quest Pro, and a PNY RTX5080 OC.

F-16 front panel 10 degree fixing quick connect adapter, and keyboard/mouse tray. The great thing about this extrusion build is that you get to "invent" all kinds of stuff to attach to it. https://blog.hempstick.org/2025/05/10-degree-quick-connect-adapter.html

Hey Fred, I am getting a CTD upon switching to VR in MSFS 2024, with the following error message in event viewer. Once I uninstall the XRFrameTool, I no longer gets the crash. Let me know what I can help to get you the dump or any info you need to track this down. I like the XRFrameTool quite a bit! I want it back! Faulting application name: FlightSimulator2024.exe, version: 1.4.20.0, time stamp: 0x00000000 Faulting module name: XR_APILAYER_FREDEMMOTT_d3d11_metrics64.dll, version: 0.2.0.139, time stamp: 0x680e99b7 Exception code: 0xc0000005 Fault offset: 0x0000000000033f84 Faulting process id: 0x9208 Faulting application start time: 0x1DBC79C4DE92F20 Faulting application path: C:\Program Files\WindowsApps\Microsoft.Limitless_1.4.20.0_x64__8wekyb3d8bbwe\FlightSimulator2024.exe Faulting module path: C:\Program Files\XRFrameTools\lib\XR_APILAYER_FREDEMMOTT_d3d11_metrics64.dll Report Id: 213bbfaa-26ef-4f64-86ec-6598ee3354f9 Faulting package full name: Microsoft.Limitless_1.4.20.0_x64__8wekyb3d8bbwe Faulting package-relative application ID: App