Headwarp

HOw hamfisted did you get? How hard did you pull and did you hear any cracking before you realized the pci-e slot had a lock? Looking at a pic of your board, you might be able to get cable that lets you plug your gpu into the bottom pci-e slot as it doesn't look like a 1070 would fit down there. It should be capable of running x16. You'd have to figure out a solution for mounting the graphics card if by chance you damaged the top pci-e slot but it should work. Something like this https://www.newegg.com/Product/Product.aspx?Item=N82E16812183053&Description=pci%20express%20cable&ignorebbr=1&cm_re=pci_express_cable-_-12-183-053-_-Product https://www.newegg.com/Product/Product.aspx?Item=9SIAGYY7NN9127&Description=vertical%20gpu&cm_re=vertical_gpu-_-9SIAGYY7NN9127-_-Product Perhaps not those specific items.. they were just the first i stumbled upon. And if you do determine the PCI-e slot is damaged, if you purchased some kind of protection plan from the retailer they might cover it, or if still under warranty by chance asrock might overlook it if you don't tell the whole truth or if your conscience demands, they might be cool with you about it but no guarantee there. Also just to be sure, you only have one monitor or display plugged into the thing right now right? it could be confused on which is primary if more than one. I can't even get into bios while my VR headset is plugged in. I'm doubtful this is your reason given if it were another monitor it would provide a picture. Provided your motherboard and fans and everything else are powering on as per normal and posting.

This is what my ram and page file usage look like when DCS is running.. I'd honestly love to see a comparison with someone using 32gb of ram. Looking at all those hard faults I'm guessing I should go on a ramen diet to get a 32 gb kit sooner than later.

Sitting here looking at your mobo's specs, says max 64GB DDR3 2400.. but looking at the qvl list there's maybe one 24gb listing and the rest are 2-16GB >.< Not quite sure I get that.. however - 32GB kits of ddr3 1600-2400mhz are listed around $200 @ Newegg. I honestly can't speak for what improvements you might see because of this but, it's a lot cheaper than a whole new build, and ram has a 30 day return policy if it doesn't offer you any noticeable improvement. If you're maintaining 45fps in VR you're pretty much golden. No amount of hardware is going to get around that for DCS at the moment. Planning on upgrading to 32GB asap myself but not soon enough to let you know what kind of benefits I might see from it.

3 year manufacters warranty I think is pretty standard. I know that's the case for both EVGA and nVidia, might want to check up on Asus/Gigabyte/MSI, but I'd imagine it's similar. The effect is honestly minimal gains in minimum framerate and playing in VR im generally stuck with motion reprojection on in DCS anyways. My 980Ti didn't even like +10mhz on the core clock so this is really the first I've ever OC'd a gpu. But, nvidia scanner gave me a good baseline to work with on the core clock and this particular sample played so nicely with my increases that it wouldn't make sense to not take advantage of even the slight boost in performance. It was pretty easy to do. So if you don't plan to OC and aren't willing to shell out $1350 for higher stock clocks, the FE models should be right up there with the competition this generation. EVGA's manufacturer's warranty is slightly better than nvidia's imo, depending on if you buy from a retailer that offers a return period for refund. EVGA will send out a brand new OEM card if there's a fault within the first 30 days of ownership, afterwhich they might send out a refurb instead. Nvidia doesn't even offer the first 30 days afaict before stating you might get a refurb. So, if your retailer doesn't offer a good return policy or has a restocking fee that's a consideration. EVGA also gives the option purchase an extended warranty for an additional 2 years, or 7 years on top of the original 3 year warranty. And you have 90 days from the original purchase date to decide whether you want to purchase said extended warranty. So with the FE at $1199, and the XC Ultra @ $1250, it kind of just boils down to whether or not the peace of mind of 24 hour tech support and a solid RMA policy is worth the extra $50 imo. I can't speak for how long an nvidia rma might take vs EVGA. But I generally buy EVGA with confidence. Asus, i've read some horror stories about rma experiences, but no personal experience, as my asus motherboards generally just tend to work. Can't speak for MSI or Gigabyte either, but it's probably pretty easy to look up their warranty terms. In the long run though, you hope you never have to use the warranty at all. So whatever you get I wish you good luck. You should have the info you need to make a choice if you're determined to get a new gpu. :)

No apologies needed. I'm thankful you've given the thread any attention.

We've established that I might have missed the FFI showing more than 15000 pph in a previous patch. Still.. I don't see patch notes about it unless I overlooked them. Thanks though. Thanks :) I couldn't tell you one way or the other, just something that caught my eye. If the IFEI fuel flow indicator in game is correct at least you guys know to update the manual regarding it. :) Not a huge deal.

I admittedly wouldn't be sure when exactly this changed, and could have just not been looking the last time I flew the hornet, as it's not my main bird when I'm flying MP for the time being. Will have to go back and see if i missed something in the logs which would clear up whether this is meant to be or not. *edit* Looking through the official thread on updates for Openbeta, unless I'm overlooking something the only mention I see that might apply to this is in the latest patch notes, "Known Issues The cockpit gauges may be broken. Will fix ASAP." It was one of the first things I noticed when I jumped into the hornet after downloading the patch, so while I can't say for sure which of the previous updates introduced this I'm still wondering if this is "working as intended." or not. Not to make a big deal or anything, but just to point it out if something's wrong.

From the DCS F/A-18C Hornet manual • Engine Fuel Flow (FF). Displays main engine fuel flow only (afterburner fuel flow is not displayed). Range is 300 to 15000 Pounds Per Hour (PPH) with 100 pound per hour increments. The tens of units positions have fixed zeros. When fuel flow is less than 320 PPH, zero is displayed. Prior to today's patch the IFEI never went above 150 per engine (or 15000 PPH) with max afterburner. Today I'm seeing them get up to 383 per engine (or 38300 PPH per engine) with max afterburner while flying just above the ground. Unsure if this is a bug, or if for some reason afterburner fuel flow is now being displayed on the IFEI and that part of the manual was a placeholder. anybody?

In response to the OP - there might be only one thing really to consider when looking at buying nvidia brand or some other brand without taking into consideration overclocking and that's the RMA process. IMO EVGA can't be beat when it comes to customer service in the hardware manufacturing business. The ONE time i had to rma a card I had my replacement within a week and a half, pretty much no questions asked. Thats from the time I shipped them the DOA card, to the time I got a new one, living on the other side of the country. That was the card I upgraded from. Other than that models like the Strix from Asus and the FTW3 from EVGA are using custom PCB's that have switchable bios and perhaps other uniquities, like the FTW3 having a 150% power target with 19 power phases if you are looking for ultra enthusiast level extremes in overclocking. (This equates to like 373w draw >.<) The FTW 3 also includes temperature sensors for VRMs, memory, and pretty much every component on the PCB, where most cards just tell you about the gpu. That being said - EVGA in their forums has a download link for an updated bios for XC Ultra users..which increases the power target from 120% to 130%, which is awesome for someone like me who bought the XC Ultra and decided to OC. Which is another factor separating an AIB card from the FE model with 120% (maybe 122%)? power target. I'm running +104 mhz on core clock and +607mhz on the memory clock which = +1214mhz effective memory OC. So 1754 boost clock, which actually gets me up to like 2050-2070 boost when gaming, starts to throttle slightly because it hits 60C with 100% fanspeed, which I may eventually watercool my card to prevent that. Completely stable, even in DCS. I would have rounded off to +105 and +600 but the sliders in EVGA Precision x1 are hard to be "precise" with. But I didn't touch the voltage slider, just the power target slider. There may be room to push the core clock further, but I'm comfortable here. The FE 2080Ti is closer to AIB cards than ever though.. if you're okay with how they handle warranty issues should something go wrong. Which, looking at the geforce forums FE models seem to be having more issues than other brands. How much of that has to do with the end user is up for speculation in my book. But out of the box, there's only about 15mhz boost clock difference between the FE and EVGA XC models, between 15-30mhz boost clock with certain Asus models. The FTW3 is in a league of its own @ $1349USD and EVGA's manhandling of the turing gpu. YMMV on the OC, but I feel like I got a nice overclocker myself. Not regretting the choice coming from a 980Ti at all given my gpu hungry monitor and VR headset. Although I almost bit the bullet on a 1080Ti and a 32gb kit of ram >.< and couldn't with the 2080Ti's pricetag. But I'll get the ram eventually anyway.

I think it feels on the dim side as well, at least in my odyssey.

You can always deactivate before switching components and reactivate

So apparently the power management option in nv control panel for my dcs profile was set to optimal power, which equates to me having a brainfart because I could've sworn it was on max performance. So, not dropping below 45fps over PG map at least in single player with my current settings in VR even over Dubai..at least in my last two flights. Still getting both icon's using wmr's new motion reprojection indicator that suggest being cpu and gpu limited. Flying over normandy was the same. Might have to go pancake mode to see what my cpu/gpu are really doing as there seems to be no real way to completely disable some form of reprojection, although as FPS go up it does turn off using the older method. I'm not overly hopeful a 32GB kit of ram is going to be helpful for much, but still plan to make that move asap. Anyway - No need to follow up on this unncessary "experiment" but I might post one more time in the following months when I have a spare $300 to throw at memory.

DLSS is just the first feature of the NGX SDK, which is going to be responsible for tensor core operations. https://news.developer.nvidia.com/dlss-what-does-it-mean-for-game-developers/ Tensor cores may be capable of handling physics, and other aspects of the rendering pipeline as well as creating smarter enemy AI, adaptive cheat detection and we have NO real idea what this technology is going to be capable of over the next 5-10 years. We haven't even scratched the surface. I'm personally curious, if the 20 series is a hybrid of the best of both worlds, and that perhaps we'll be less reliant on CUDA in the future making more room for Tensor and RT cores on the die. Deep Learning AI IS the same technology being used to teach vehicles to drive themselves in the real world. It may be the next generations of cards before we see gaming applications beyond DLSS from it, but, the future of gaming could be very interesting. RTX is just paving the way. Now if they'd release some stuff for us to experience it with..given that it's November >.< I mean.. am I really the only one who thinks the possibility of a supercomputer training itself to fly AI aircraft with more human like reactions and decisions doesn't sound like amazing potential for the future of combat flight sims? It's just a pipedream right now but, this is stuff that was previously considered impossible, or at least required too much effort. Anyway - all just speculation right now. Grain of salt and all. Who knows at what point a dev team would have to invest into their own $130K supercomputer to train the neural networks required for such tasks. I'm not hopeful, just intrigued. Have been hearing about Deep Learning for awhile now, and I'd been wondering if it was going to find a place in gaming. Not going to lie.. I like the sound of Deep Learning AI for gaming more than I like the sound of it for self driving cars. I mean.. ever seen the movie Maximum Overdrive? That movie was never scary to me, but now that we're actually building Skynet and cars are being taught to drive without human input i mean...

=x sorry SVSmokey =x, not monkey.

I haven't made it that far in to the Epsom campaign or flown on the Normandy map for quite some time. I'm currently recovering from hardware purchases at the moment but plan to get 32GB of 3200mhz CAS 16 ram myself. Whether this will help in Normandy campaign for the spit I can't be certain, but I do remember a time where the Su-25T Georgian Oil War campaign in DCS World 1.5 was maxing out the 8GB of ram in my system at the time and caused me much lower fps than I'd like due to the large amount of units and upgrading to 16GB provided a substantial FPS increase. You can add memory usage to MSI Afterburner's OSD, as well as page file usage, which I think is actually ram usage + page file usage. http://www.tomshardware.com/answers/id-3553452/ram-speed-affect-gaming.html Some comparisons about speed of ram. IMO, 3200mhz is about the best price to performance ratio when sticking to 16 CAS. I'm currently running into either a CPU or memory bottleneck myself in the Dubai region of the Persian gulf map in VR that I'm hoping a ram upgrade alleviates. Good luck with your GPU upgrade SVSMonkey. Just bare in mind changing one piece of hardware can run you into a bottleneck on another component. Hope it gets you the 15% you were after.

hehe I love how you use the word "Conservative" to describe that kind of memory clockspeed increase. Which my 2080Ti doesn't seem to mind either. But, that's 15200Mhz effective right there.

The price of cas 16 3200mhz ram is honestly not too far off from the fastest 2133mhz sticks, at least on Newegg. CAS 15 and CAS 14 start to get up there in price but personally I'd just get the cas 16 32 GB 3200mhz as the price difference is maybe like $40 USD over many 32GB 2133mhz kits afaict. If anything the extra $40 is worth not driving yourself crazy over choices. :) As long as your mobo supports that speed.

Make sure your power options in windows are set to high performance. If using windows 10 type "Control panel" in cortana and hit enter, double click "Power options", by default it should be set to "Balanced" but you want the "High performance" option, which may be under "additional plans" which may require clicking a drop down box to show the option. I think that SHOULD be all that's required to keep your PC running at 3.6ghz when running DCS, but at this point if not - check temps or, it could be the fact that your machine is using more than 1-2 cores, but grain of salt there as I don't know if that's how it works on 6th gen processors. On newer processers, Turbo does two things, 8700k for example, it will either boost 1 or 2 cores to 4.7ghz by default, or all cores to 4.3ghz depending on CPU load. To get around that requires overclocking, and in my case disabling the AVX offset, which was necessary as both of my favorite combat sims were clocking me @4.3ghz despite being heavily dependent on single threaded performance prior to my decision to just overclock all cores. Unsure how that i5 6500 handles it. It either boost all cores or just one or two cores. But definitely check your windows power options are set to high performance, and next time you're shopping intel CPU's definitely go for a K series for flexibility.

Thanks for answering this guy's post Nineline. I'll also take this opportunity to thank you guys for all the exciting news this year from 2.5 and the Hornet, to dedi's and even wag's youtube comment about actively working on a dynamic campaign, as well as the missile rework. Feel like you and the team have also been far more communicative as well and that may just be me, but I think it looks good for the future of DCS.

Just glancing over your thread, I've been tinkering around over PG in the huey and the Dubai area seems to really be hungry for resources. I was just using MSI Aftburner and lifting my headset to check, and at times I was running into CPU limitations, however my GPU was still bouncing between 97-99% and this doesn't seem to change if I'm using 1.5 pixel density or 2.0 pixel density, which you would think the added resolution would put the workload on the GPU and lessen the load on the CPU. Today, I enabled WMR's new EMR (new motion reprojection) and the reprojection indicator which shows a different color square depending on the state of motion reprojection - green = able to play @ 90FPS, light blue = on due to cpu limitation, dark blue - on due to gpu limitation, red - off because can't maintain 45fps. So after turning all of that on, I went and flew around caucasus with the dark blue square meaning on because gpu limited. Then loaded up city tour instant action for the huey over Persian gulf which starts you in Dubai.. the entire time I was flying over the city I had both a light blue and a dark blue square, indicating I was maxing out my CPU and GPU. SO I started flying towards the eastern coast away from the city and eventually had only the dark blue square, but when I turned my head and looked at the city both squares were popping up. Ultimately it's like I'm at some kind of sweet spot specific to Dubai in the Persian Gulf map where my gpu is still being fully loaded by VR but running into CPU limitations which is what I suspect is causing the framedrops. Fragbum's also getting fps drops, but has 64GB of ram, so now I'm wondering if you're seeing spikes to 99% on any of your cpu cores while flying over Dubai. It might be happening so briefly that you need to lower your hardware monitor's polling rate but I'm curious what's causing CPU usage over dubai to be so much higher than anywhere else on the map. If it's just the big buildings or if we're encountering some kind of shader issue that needs to be optimized. If fragbum isn't seeing any cpu limitations I could very well be choking on 16gb of ram. In comparison - flying over caucasus it's always the dark blue square and no single core on the cpu ever reaches 100%. But I mean.. I'm on an 8700k @ 4.9ghz, to run into cpu limitations with up to 2x supersampling in the odyssey just seems weird.

A big difference maybe, but in my personal experience from switching from a sata III SSD to a 970evo for boot, is that I hardly notice a difference if at all. Sure, my storage benchmarks are through the roof on that particular drive, but in the actual application of boot time from logo to windows, I was still in the same scenario where if I wasn't paying attention I'd look up and find myself happily surprised at how quickly I'd load into windows vs a sata III SSD. Maybe it's just me but I'm not sitting there with a stopwatch timing it every time I restart. Booting from either drive type I hardly have time to glance at my phone before being at the login screen. And really how often am I actually rebooting my system? There could be some truth to the nvme handling page files better, but it all depends on what's contained in the page file. Large files? Yes, batches of small files? Not so much. Even then a better yet answer to paging files is an increase in DRAM. I stand by my statement, it's hard to go wrong when it comes to SSD drives. No sense complicating the process. If you aren't interested in saving a buck, splurge on the nvme drives, if more space for your dollar is your concern, get the sata III ssd. Either will be sufficient purely for the purpose of booting to windows or running DCS imo. If we were talking about "Do I want an HDD or an SSD" I'd probably treat it as a bigger deal but aside from transfering data, once you've eliminated the mechanical action of spinning magnetic discs, small file access times are hardly utilizing the full brunt of even the slowest SSDs.

I'll have to record a video in flight for comparison. I only really saw the shimmering in your vid on console and instrument panels and then they seemed to sharpen up as you were in the air. I will say that no matter what I do in my Odyssey there is some degree of shimmer in DCS and I'm more inclined to think that has to do with DCS's texture resolution honestly and msaa tends to reduce the effect. Or at least used to. Who knows what can happen with any given patch. In your circumstance with it not being specific to DCS though I'm at a loss.

Mustang replied to my post before I could come back and delete it. :) Which is on me I should have tested more before posting in the first place. Not exactly sure if my issue is related to DRAM. But fixed pagefile didn't fix what I was trying to fix in the long run. Going from 16 to 32GB is about the only real room for an upgrade at the moment with my rig. Which is the game plan eventually anyway. My wallet's just had a lot of abuse this year at the expense of PC hardware >.<

I forget which hardware reviewer I watched, but for the SSD's they tested, about 40-60 seconds to boot on an HDD, 11 seconds to boot on a sata III SSD, and 6 seconds to boot on an NVME drive. And I believe they were timing to the point that windows was fully loaded after logging in. My point was simply that either type of SSD is far superior to an HDD in terms of performance and load times, where the gap between sata 3 and nvme for boot speeds isn't quite so large.

Kind of a necro, but just unchecked the "Automatically manage paging file size for all drives" checkbox, and set up a fixed 32GB pagefile on my C: drive, because it's my fastest drive being NVME, and so far it seems to eliminate some occasional hiccups I've been encountering lately. (Hindsight I think I hadn't looked at my pagefile settings since my latest windows install, either that or a windows update reset it to default) When I unchecked it, it showed system managed on C: and none on all my other drives, but I'm wondering if the automatic management was for some reason using my HDD or something. Also curious if it was just the pagefile adjusting its size on demand causing it, as another poster in this thread mentioned. Will have to give it some more testing, using "system managed" on C: but without the checkbox for automatic paging file management. First, just need more stick time to verify that this really did something. Which would be nice, because I wasn't really looking forward to shelling out for a 32GB kit of ram.