Jump to content

Voyager

Members
  • Content Count

    568
  • Joined

  • Last visited

Community Reputation

174 Excellent

1 Follower

About Voyager

  • Rank
    Member

Recent Profile Visitors

870 profile views
  1. @Gambit21 @Jaws2002 The expectation is that nVidia is going to do a 20GB 3080 Ti, which would resolve the memory issue, but I'd expect it to be $1k USD, so we'll have to see how the 6900 XT performs. Also, the 6000 series may turn out to have a decisive advantage or disadvantage for flight simulator type applications, so I really want to see how it handles FS2020.
  2. What's interesting is that the Radeon cards seem to have more stable frametimes than the 30 series. I'm curious how that will pan out in VR. While it does seem like the 30 series do handle 4k+ slightly better than the 6000 series, the frame time stability may turn out to be a bigger impact in practice. I also notice that the 2.5Ghz OCs seem to beostly power and temperature limited, so while I don't think we'll be able to reliably hit anywhere near that in the stock cards, it really does open up the possibilities for the AIB cards.
  3. That would be handy. I could clamp it on my desk, loop the cables in it, hang the headset on it, and if there's enough reach, even hang my headphones on it, or if there isn't, I could use it for the cables and headphones, and tack something like this to the top of my tower and keep the VR headset there: Amazon.com: AMVR Upgraded Version 2nd VR Stand,More Stable Base Headset Display Holder and Controller Mount Station for Oculus Quest, Quest 2, Rift, Rift S Headset and Touch Controllers (oculus-stand-V2nd): Electronics https://www.amazon.com/AMVR-Upgraded-Controller-Controllers-oculus-stand-V2nd/dp/B082F2XTD9/ref=mp_s_a_1_15?dchild=1&keywords=vr+headset+stand&qid=1605636035&sprefix=Vr+headset+sta&sr=8-15 @Goffik We've got high ceilings in here, and I'm next to a window, so I don't have any goo spots to wall or ceiling mount it here, but I can see that working with other geometries. I've actually been looking into ways I could set up my entire VR/Centerstick rig as a separate unit from my main desktop, simply so I can leave it set up and ready to fly between gaming sessions, but haven't figured out any geometry that would work yet. 😕 maybe if I turned my desk towards the window and set up the VR kit behind me? Though the problem has been getting the throttle in a good spot, and that is movable, so maybe I should simply embrace that? Project for this weekend, I think...
  4. Was wondering, does anyone know of any good VR headset stands with something to hanger the cables on? I'm using an HP Reverb. Right now, when I'm not using it, it lives on top of my tower, but that's both, not hugely stable, and all that cable just want to go under my chair. I'm thinking there has to be a better way to set it up. Looking through Amazon listings, I'm seeing a lot of PSVR headset stands, and a number of headset shields for the Quest, but not much with anything for corraling the piles of cables. Thoughts? Thank you, Harry Voyager
  5. One question: have we confirmed that the Zen 3 chips are measurably superior to the 10900k parts? The arrival of Zen 3 has put Intel on the back foot, so at my local Microcenter, they are selling 10900K's for $540 USD, and they have them in stock. I'm already on a AM4 platform, so a Zen 3 part makes the most sense for me, but if the 10900K and 5800X are comparable in performance for Il-2, and are now the same price, about, and you need to buy a new motherboard and ram, it may be worth considering.
  6. @Jaws2002 It looks like the 5600X and 5800X chips could hit the 5950X numbers, but they're running into the hardwired boost cap instead.
  7. Very cool. Just saw that the local Microcenter still has a 5800X in stock, but we both should rebuild fund for a bit (had to replace a chunk of the AC last week) and I kind of want to go with the 5950X, so I (in theory) don't have to replace the CPU on this thing for a fairly long time. With Il-2 and DCS being the only real things driving me to an upgrade, I am extremely interested in what the tests turn up.
  8. @Gambit21 At least according to the Anandtech review, the Zen 3 has much improved core to core latency within the CCX, and apparently is better than Intel's now, so long as you don't jump CCX: Core-to-Core Latency and Cache Performance - AMD Zen 3 Ryzen Deep Dive Review: 5950X, 5900X, 5800X and 5600X Tested https://www.anandtech.com/show/16214/amd-zen-3-ryzen-deep-dive-review-5950x-5900x-5800x-and-5700x-tested/5 If the core to core latency is the root cause, or a major contributor to the Zen 2 performance deficit, we should see the 5800X end up out performing the 5900X in Il-2.
  9. Well, at least according to the HW Unboxed 5950X review the 5950X did beat the 10900K in Squadrons, another DX11 flight like title @13:46 while the 3950X did not. I don't know if that will be relevant or not, but it could be a useful indicator. I'm really looking forward to our forum benchmarks here.
  10. Trying to get to the root of the performance difference between Intel and AMD CPUs, especially vs other game genres, and suspect there may be a polygon count difference in flight sims. (Especially since it seems to be New York City that's the CPU sledgehammer to beat all CPU sledgehammers.) I'm wondering if there is a way to measure or get a report out of the number of displayed polygons in a given IL-2 scene or graphics setting? Main thing here is pretty soon we're going to have three different CPU architectures to decide between (Zen 3, Sky Lake, Rocket Lake) and I don't really know what it is that's making Sky Lake work best for flight games. Is it just the frequency, or is there something else specific about it, that may or may not be there for Rocket Lake or Zen 3. Thank you, Harry Voyager
  11. @robbiec What's striking is your 3900 is getting only 1.9m calls, while my 3800 is getting 2.2m: http://www.3dmark.com/aot/378472 That is even though your chip is hitting 4.67Ghz during the test while mine is only holding 4.45Ghz I wonder what it would take to get the 3DMark folks to let this data be seachable? Right now it sucks that we have to post direct links to specific results.
  12. I thought they'd both openness up their standards such that both now worked with both sync systems?
  13. To blow hot and cold with the same breath, we'll want to be very aware of the performance at a given resolution. It does sound like at 4K+ resolutions, the 30 series performs better. It could easily be that whether we want a 30 or a 6k depends entirely on what resolution we're running at. While at the moment, I'm leaning towards a 6900, if a 3080 TI comes out with the 384 bit bus, that might, for me, be the better option in VR. (4k is 8m pixels, but the Reverb is 9.3m pixels. 1440 Ultra wide is only 4.8m pixels) I am *really* looking forward to the November bemchmarks.
  14. I was getting about 2 278 275 on my 3800X at a reported 4.45Ghz, but the 3800X is a single CCX chip, unlike the 3900. 3DMark API Overhead Results: Are API Draw calls the limiting factor for VR performance? - Virtual Reality and VR Controllers - IL-2 Sturmovik Forum https://forum.il2sturmovik.com/topic/66124-3dmark-api-overhead-results-are-api-draw-calls-the-limiting-factor-for-vr-performance/ If this is the bottleneck, it looks like it is intrinsic to Dx11, rather than being Il-2 specific. Remember DCS and FS2020 are both seeing this same sort of difference, and they are both on completely unrelated engines. I think the only big similarities are both are DX11 Flight sims.
×
×
  • Create New...