RGS

Members
  • Content Count

    54
  • Joined

  • Last visited

Community Reputation

1 Neutral

Flight Sim Profile

  • Commercial Member
    No
  • Online Flight Organization Membership
    none
  • Virtual Airlines
    No
  1. Wow, that's a pretty substantial OC. I think the 9900K can run 2 cores at 5 GHz, with all 8 at 4.7. Is the sim (P3D) making use of those extra cores on your 7920X? How do you find the processor outside of flight sim/gaming? Thanks.
  2. Thanks for the reply, Westman. Do you happen to know if that's the case at 4K as well? The 9920X is a slight step up from the 7920X, but I imagine not enough to make a significant difference. If I was just considering a machine for flight sim/gaming with a thought to changing it every 2-4 years I'd just go with the 9900K + 32 GB RAM and be done with it, but as I have to take into account the other uses for the system, namely Max and Photoshop and the fact that I'll be sticking with it far beyond that time frame (graphics card upgrades aside) things become less obvious. Totally get that right now the 9900K is the best gaming processor available; my dilemma however is, is it significantly faster than the 9920X in flight sims/games, and in particular at 4K (plus will that continue to be the case in the years ahead); also how much of an advantage would the additional 4 cores and quad channel memory give me regarding the work side of things? In some ways I suspect that the higher clock speeds of the 9900K may actually make it the faster of the two in Max and Photoshop (barring rendering), but not 100% on that front. As generally speaking I'm not rendering video from Max (only stills, lighting + normal map bakes), rendering time isn't as important a consideration as it might otherwise be. The 9900K is tempting, as the far cheaper option; my concern is whether it will it last as well as the 12 core part in the long run? I don't want to overclock as 1: I don't know what I'm doing 😉, 2: I need stability and don't want to spend a lot of time troubleshooting etc. and 3: component longevity is important to me. Sorry for the long-ish posts. Thanks for the feedback.
  3. Hi all, Looking for some advice on a new system and wondering which of the two above processors to build it around. My primary uses: Flight sim (P3D), obviously... Games 3ds Max (primarily model creation, but also rendering) Photoshop 4K All of the above are of roughly equal importance and are kind of linked, as I'm a developer. In addition, I'd like to add video editing, though that's of less importance right now. I do work with very large files and often have multiple apps open (Max and Photoshop for e.g.). My intention is to pair the system with an RTX 2080 Ti. Whilst expensive, it does deliver the goods at 4K. A few further points to bear in mind: This is my primary work machine and will see heavy usage for years to come I do not want to have to upgrade the processor/motherboard, but am OK with mid-life GPU + RAM upgrades (but ideally would not want to for at least 3 years) I will not be overclocking I will only ever be running a single graphics card System must be Intel/nVidia and be as future-proof as possible (I tend to stick with a machine for much longer than most, rather than frequent upgrades, beyond the graphics card) Price differential between the two systems is currently around £700-800, which is considerable (roughly £2,500 vs £3,200 for the entire rig, minus display) As I understand it the 9900K is faster at 5 GHz vs 4.5, but the 9920X has half as many cores again (12 vs 8) + supports quad channel memory, though not sure how important that is in terms of real world usage however. Benchmarks seem to show the 9900K producing higher framerates for games at lower resolutions (due to the higher clocks), but things flatten out at 4K. With that in mind, I'm tempted towards the 12 core part which ought to help with work (rendering if little else) and potentially be more future-proof as apps eventually embrace higher core counts (though a very slow process, if the past few years are anything to go by). On the other hand, 5 vs 4.5 GHz is a considerable margin in terms of raw speed,and those 4K figures may shift in the direction of the 9900K as games become more demanding. The 32 vs 64 GB RAM question is more to do with future-proofing. 32 is ample for now, work-wise, but in the past I've been burnt with RAM becoming discontinued and/or insanely expensive, so am considering just going the whole hog now. Would be interested to hear from others, particularly with regards to large files in Photoshop, heavy Max usage and video editing. I don't imagine flight sims/games will be taking advantage of over 32 GB for some time, though maybe 4+ years from now, who knows? Thanks in advance for the help. Cheers!
  4. RGS

    Jordan King - Prepar3D v4 First Look

    I'd imagine LM need to write the rain effect shader first (and 3ds Max tools to go with it) - that's their job. Once the system is in place, devs then have the tools/control required to correctly implement the effects of a per a/c basis, flagging the appropriate polygons, setting wiper masks, parameters etc. Disappointed not to see the base effect added by LM at this stage, seems like a very odd thing to leave out when going to the trouble of re-doing the rain and snow, particularly when FSW now has the feature (and it can't be that hard to implement).
  5. Good topic. I'm particularly interested in the impact that going to a bigger screen might have on one's vision. I've always used large monitors, from 21 inch 4:3 CRTs (back in the day!) to 2x 24 inch 16:10 LCDs and for the past 8 or so years a 30 inch Apple Cinema Display (I'm not a fan of multiple monitors; the 2x setup was prior to working for myself). The 30 inch ACD is by far the best monitor I've owned to date and has given me less eye strain than any used previously (with the CRTs being the worst, particularly as they aged). I would have thought that a larger screen could actually be a lot better for your eyes than a smaller one as: 1. You tend to sit further away from it. 2. Your eyes actually have to move *a lot* more, darting from corner to corner (compare this to reading on a phone/tablet at close range). Ideally I'd like to switch to a 55 inch 4K 60Hz Sony TV, for both full-time work (3D/2D graphics/design primarily) and simming/gaming (I remember the upgrade to the 30 inch being a real eye opener; it felt as though I was no longer looking through a window, but now it feels a little limited and 4K offers the chance to go bigger without compromising the details). I view my current monitor at arm's length; the TV would be about double that distance. Due to the large size I'd be moving my eyes even more than the 30 inch, which I'd have thought would be a good thing, but am slightly concerned about sitting in front of such a beast for such extended time periods, particularly as it's a TV and not a monitor (not sure how much of a difference there is these days, but obviously one was designed for extended use at close-ish range, the other not). Luckily I sit next to a window with a great view, so giving my eyes a rest every now and then is easy . Thanks for your thoughts, Robert
  6. RGS

    SpeedTree? Where?

    Is it possible to still download + run MS Flight? I'm interested to look at full SpeedTree integration + also see what they did with some other effects such as sun glare etc. I did give it a test at time of release, but due to its arcade/casual nature quickly dismissed it. Cheers, Robert
  7. RGS

    SpeedTree? Where?

    Yeah - I really hope so. That'd be excellent .
  8. RGS

    Speed Trees

    LODs should take care of the worst of the performance worries (obviously there'd still be a hit) and instancing help address the memory issues. It's very important to have an consistent world IMO; a mix of high quality and low quality visuals is a real enemy of immersion. These trees, applied globally, would make such a massive difference to helo flight.
  9. @Brandon Thanks for the info. I reported the shadow bug on the LM forums a while ago; I understand it's a little difficult as you're effectively having to render the shadow of a hidden object (exterior model) but don't want the performance hit of having to calculate it (and it's handled differently to FSX). Still there must be a solution, it's clearly broken and shouldn't just be left as is IMO. Shame about the lack of a sun glare; unlike the shadow issue I really can't understand how this has escaped them. It's such an easy effect to get in (you could probably even buy it off the shelf!) and aside from adding greatly to the immersion is also rather important for realism and accurate training I'd have thought. I really don't understand the absence of things like this. Rain on the wind shield (not the ancient FS9 rendition...) is another one. The effect would have to be setup by third party devs for their a/c, but requires the coding, tools and shaders to be set up by LM (as they are now doing with the SpeedTree tech). Though more difficult than the sun glare, this effect (Train Sim version embedded for ref. at end of post) should be pretty easy to get working from a graphics programmer standpoint I imagine. I'm sure it must be present in a number of modern racing games too and thus shouldn't require a lot of R&D (rain running down the a/c skin would be another nice touch, though far less important). I don't mind LM leaving certain things to third party devs, in particular aircraft and high detailed sceneries (which makes total sense to me), but I really think they should strive to increase the fidelity of the base sim across the board. We've seen some great stuff with the cockpit shadows, HDR, volumetric fog and 3D waves (and I'm really looking forward to the avatar mode) but there are some simple things, which would not be difficult to introduce and provide an instant 'win' that just seem to get ignored. Rain on wind shield effect:
  10. RGS

    Speed Trees

    I really hope we'll see these as an official LM replacement for all autogen trees. They're clearly a big step up from the old 3 plane ones and have a good track record in the games industry. Couldn't the models simply overwrite the old ones (on a like for like basis) without even having to touch the autogen tree placement? Whilst I totally understand that there will be an FPS hit; one of the good things about SpeedTree is the LODing system. All of the distant trees will just be quads anyway, it's only those you're close to that are high detail. Not only that, but the existing models and textures can't take up a lot of space - I imagine it'd be easy to just add a tick-box under performance settings for low detail/legacy trees. I think it'd not only be a great shame to limit these to third party airports and sceneries, but it also creates an inconsistent world filled with massive quality gulfs. As a helicopter pilot, you'd desperately want to see these new trees wherever you fly and not just specific locations. In fact, due to the inconsistency, I'd almost rather just the old quad-based models than an awkward mix of the two which impacts on immersion and the sense of a believable world. Here's hoping...
  11. @Brandon Thanks a lot for the info, much appreciated. So A and D are used for turn rather than strafe? That might feel a little on the odd-side... I'd have thought a middle mouse click or similar to toggle mouse look on/off would have done the trick, but I guess I haven't tried it yet so maybe best not to comment . Does it feel natural and smooth as a standard FPS would, or is it a little on the clunky side? Also, I added a cheeky extra Q about the sun glare (or lack thereof) above if you have time to answer. Thanks again. (if there are any extra graphical tweaks they've made that have impressed you I'd love to hear them!)
  12. One more quick Q for you beta guys: Does the aircraft's shadow cast on the terrain now correctly reflect that of the exterior model or is it still the VC? EDIT: Actually, make that two Qs Is there a new sun glare effect? I notice that 'removed old sun glare' was listed in the log, but in Rob's most recent video there doesn't seem to be any glare at all. Cheers. Am pretty sure it's been confirmed (via Orbx) there are no new trees actually included in the sim as such, only the separate models and the tools with which they can now be added by third party devs. Bit of a shame really; upon reading the change-log I'd hoped they'd updated the base autogen. Still, it's a nice feature for further down the line.
  13. @Brandon How does the avatar mode work? Is it simply another viewpoint you select (and then first of third person) or is there more to it than that? I assume controls are via keyboard and mouse as with a standard FPS? Also, do you happen to know whether if the a/c is developed with the avatar in mind you can use the same model as the pilot? Thanks for the info.
  14. ATC window looks great. Really nice; clean and minimal just how I like a UI .
  15. Anyone else get the impression that v.4 may well not be too far behind v.3 (in terms of release)? Firstly, a few testers commented that they were aware of both versions 3 and 4. Then earlier today someone was asking about increasing the LOD distance on the P3D forum. The first response was from a tester who remarked that the feature wasn't in v3, but would/should be in v4 and the second comment, from a moderator, was to read the previous reply and "Watch this space!" These posts have subsequently been removed by Beau who stated fairly enough that "I deleted the posts referencing v4 to avoid confusion and edited my post for clarity. We aren't currently commenting on the timing or content of future major/minor releases." "Watch this space" doesn't sound like the sort of thing you'd say for a wait of a year or two (I'd have thought anyway), making me wonder if v.4 could effectively be v.3 in terms of features/content, but 64-bit and potentially Dx 12... Or maybe LM have simply decided to release a new version each year and cut down on the point releases . Elsewhere you can read comments such as "Add-on content with paths configured for v3 will update much more seamlessly with future releases of Prepar3d" which also suggest to me that LM are thinking about v.4 and not just 3.x which strikes me as a little unusual given the previous versions' two year life cycles. As to the cost of the upgrade/new version: I suppose it depends how serious you are about flight sim. £120 is a lot if you view it in 'game' terms and the content you get is IMO less than you would have between previous versions of Microsoft FS (for e.g. there has been no new terrain textures, autogen or clouds in any of the releases and very little in the way of 'designed for the new version' or even first party aircraft). In contrast, I would expect Dovetail's 2016 offering (or FS 11 had it been made) to have all new graphics, whereas LM are still using much of what was there with FSX. I'd expect a new MFS/more mainstream title release to have fully implemented the SpeedTree tech as autogen for e.g. and not just 'unlocked' it (together with all new autogen buildings, 3D grass and normal map support for terrain textures - Yes, there'd be a performance hit, but the low detail SpeedTree LODs would just be planes in the distance anyway and I imagine it wouldn't have been all that difficult to maintain the legacy models as a low-end/tick-box option). I've not yet seen v.3 in the flesh myself, so it's hard to comment, but certain graphical elements such as a decent sun glare and rain effect (particularly on the wind shield) are oddly MIA for a 'full title' along with other bits and pieces such as built-in real world weather (as we had with FSX). I get the impression that P3D's budget is nowhere near that of previous MFS releases though, and rightly or wrongly LM leave a lot of base sim graphics to third party developers (REX + Orbx for e.g.). Of course I might be completely out on that one; clearly the product's got an entirely different focus and primary target audience. The actual release itself is also incredibly low-key; no previews or high quality screenshots even, just a few links to some previously released YouTube vids made by testers, which many had just assumed were 2.5, and aside from the avatar one (yes!) showed nothing new at all - Let's face it, if you have to ask "Is this the new version?" it doesn't exactly point towards monumental changes. It's probably not the case, but to me this seemingly casual launch gives the impression of a small development team working with limited resources. Again though, this is just the impression I get if ignoring the different target audience etc, etc... That said, given the state of the simulation industry I for one am very glad LM are there to take care of the serious side of things, even if releases are a little light in the way of content at times. DCS is a great sim, though purely combat oriented and somewhat limited outside of that scope and who knows what DTG will bring to the table in 2016 with their new sim (other than a lot of DLC of course ). By it's very nature and license P3D is in no danger whatsoever of being 'dumbed down' and if it wasn't for LM we'd still all be using the ageing and no longer supported FSX (which, good as it is, is looking a bit dated these days). P3D gives us all the familiarity of FSX, but with a host of bonuses and is under constant development. For the serious simmer, £120 across 2 years for an updated core sim (which only LM themselves can deliver) is absolutely worth it IMO. The VC shadows and HDR alone breathed a huge amount of life into the platform and made it feel current again and the 3D waves were also a great addition. I really hope we see some new graphical features in 3.0 and more to come in subsequent 3.x releases. There are certainly a few new things for designers to play with at launch (SpeedTree and real-time reflections to name a couple) which is excellent news, even if LM haven't given any examples of what's achievable through their own content. I'm also personally pretty pumped about the avatar system (first person will be my pref.) as aside from a/c inspection, I think it'll help to ground the world around you and should make the experience much more tangible and immersive. So... Very much looking forward to v.3 myself and seeing what's possible with the platform over the coming months. I do wish that LM would be a bit more communicative about their plans though and that v.4 won't arrive in 4-6 months time for £120 without an upgrade path .