Jump to content

Gryzor

Members
  • Posts

    1221
  • Joined

  • Last visited

Everything posted by Gryzor

  1. Seriously ED: look VR implementation from other IIWW sim from "the other company", and take it as a example or a point of reference (in VR, I´m referring, not other things); The VR implementation/improvements was not abandoned..., just the opposite: huge improvement in clarity of terrain and planes - sharpness tweaks that allows VR users to have great detail, not a blurry mess -, visibility of other planes (and sun reflections from distance, that helps awareness and dogfighting without "hacking labels" or mod label parameters, just we have to hack in DCS...), the improved lighting, the deferred texturing ALSO, even parallax occlusion.... those were SIGNIFICANT changes. All of them, with many types of shadows that doesn´t cut performance (even shadows that covers even individual trees, in VR). That SIM improved a lot the past year. Even the helix-disc from 2WW planes is implement now without graphical glitches (they made a fix specific to VR, when ASW or motion smoothing quicks in). In DCS, IIGM planes like SPITFIRE due ASW in VR is nearly unplayable due the helix graphical glitches (and yes, it is the responsability of third parties, but is your base product, and you´re selling it). Those graphical improvements "from the other sim" are SIGNIFICANT changes... and most times, mantaining 90 fps during battles, without single digit less in performance. Not just a shader tweaking with incresed bloom and pallet colours, like in DCS, that bring many many glitches in 3D graphics. I´m sure that most of us are aware of those changes from "the other sim". The only VR improvement in the past year: just a shader MOD by a forum user, a single persom did much more that a complete team of programmers and coders. Simply, that is not acceptable. Summarizing the situation, in my historical simming experience: One yer ago, in 2018, the % of my SIM time were 90%DCS and only 10% the other sim / Now, in 2020, its just the opposite. Due gliches and fps, I only fly DCS just to test new beta and not much more. Take into account that in this community, many of us fly in VR, just because, like me, I´m sure that combat flight simulation is our PASSION since we were kids, and we want the best experience that could be achievable. ED interest in VR community seems to me completely abandoned, I hope some inflexion point in the near future, please take this as a the only PRIORITY. If not, I´m sure I won´t buy no other modules from ED, saying in reddit that most fixes are delayed due the "spagetti code" is not a excuse. Its your product that you´re selling, its your responsability. This is the last time for me.
  2. Maybe less GPU memory could be related to stutters in caucasus preloading, due not preload terrain (using tgp or sensers is very easy to see it).
  3. I doubt lighting as a cause of performance. A lot of people post many tests even with different settings, many of them, a common point is unusual cpu usage. With mostly high end computers here in VR. Whats the point in implementing carrier ops? to have single digit stuttering experience? to quit from VR passing to 2D? there is no CPU/GPU on this planet that could move this, with any low or high graphics settings. In the past I see some significant changes like DCS 1.5 version, more than now from 2.5.5 to 2.5.6. Let me remember to you: 1-. With new edge engine implementation, year 2015 october --> nice performance gains, from 1.0 to 1.5, earth curvature, draw distance of terrain hugely incresad,... excellent work on 2015. Brilliant, the move to 1.5 edge engine. VERY GOOD WORK. 2-. DCS 2.0 Nevada, or the inital 2.5 version from 2018 with the new caucasus terrain, was great also: less fps, but understandable due the new terrain, deferred shading, many higher number of objects, new trees, etc...and with performance fixes from DEVs in beta in 15 or 20 days only, hotfixes,... not in 3 months like now. In addition, with many less stuttering due change in terrain version, in comparison with 1.5 or 1.0. Different loading (or preloading) terrain that give us a smoother experience. An improvement over edge engine. VERY GOOD WORK. With fixes and stable version in only 1.5 month, and hotfixes addressing performance issues. 3-. But NOW: I only see some changes in shader properties in lighting, more vivid and more contrasted, different colout properties or tones, a bit longer night lighting,... not much more. ONLY that (appart to have other new VR problems like double clouds in one eye, lighting seen in one eye, the list is huge...). But those aren´t significant changes like new terrain engine (edge), new shadows, more LOD in terrain (caucasus 2.5.0), or complete deferred shading implementation (2.0b and 2.5.1), etc... like was implemented in the past. Never I tested a Beta or Alpha (2.0) or other version from ED Devs, with the problems we have now, its a completely disaster from my point of view, not comparable and not acceptable (I understand its a beta, but due we are testers... I only would some explanation from DEV what is happening NOW). It seems to me that VR are are completely abandoned or at least, not tested, not giving any priority (in fact, there is no changes since more than 1 year in VR, but many new more bugs that 1.5 years). I don´t understand why fix "memory leaks" if we cannot play 90% of missions (missions with more than 10-20 units), please consider return to 2.5.5 if finally trere is no solution or fix to fps problem (at least, with the SAME level of 2.5.5). I only hope the message you pass to DEVs will be taken in account, because if not, I only would have the opportunity of recover my money I paid in EA carrier ops module, because with this performance in VR, with this approach, I´m sure I will not use. It will be so kind if I could have that possiblity now, to recover what I invested trusting in ED of having a more optimized product in 2020.
  4. Totally agree with you. I´m following as a loyal customer (and buying) ALL ED´s works since 1998 Flanker 2.0 times, Lomac, flaming cliffs, etc.. I have all of them. I don´t remember anything breaking the sim like 2.5.6. was. Maybe it would be a wise decission to consider to returning to 2.5.5 branch, start from zero, looking changes that impact performance, maybe increase QA checking (or at least, as a minimum, finding the item of such performance fps impact, specially on VR). Or moving urgently to priorize about other tasks like vulkan or core optimizations, sincerely ED... I regret of buying early access Carrier OPS module, because with this performance, I won´t taste it. I´m also a bit tired of some comments like "we will improve peformance this year" or so, when the reality what we have is seeing worse performance after each new version. I would like some explanation about what causes 2.5.6 downgrade in FPS, even taking into consideration is a BETA, I know.
  5. Me too. looking all the way followed thorough 3 betas since january, I expect no less than a miracle to fix this, or returning to 2.5.5 levels. At least I would like to have some explanation why such huge impact in performance, appart of the comments like "we are looking / working on it". For me, shading changes in illumination doesn´t justify huge performance impact (more related to CPU).
  6. There is no new graphical engine in 2.5.6. A new updated graphic engine must come with vulkan, sincerely I hope seeing at least vulkan implementation this year.
  7. Yes, there are many problems, one of them is common as you say: The AI units peformance impact related to units it seems to me about 1.5 / 2.0 times worse, depending of the number of units, in comparison with 2.5.5. Maybe it could be related to an increased frequency of programming with line of sight or pathfinding of AI units (for example, an increase in AI frequency "search", from ms to ns?? maybe beacuse of that the CPU cycles is increased). But in caucasus, in addition, there are huge FPS stuttering problem, that is related to use of TGP, or simply zooming in VR to the horizon (is the same principle). That issue didn´t occour in Persian Gulf (for example, you can try using Su-25T in training ground targets quick mission, in PG is playable with no stutters when using TV camera. However, in Caucausus its a stutterfest the ground attack missions, specially when TV camera looks to horizon).
  8. I have the same performance as previous 2.5.6 iterations. Caucasus its unplayable, it sames something related with high CPU usage and terrain loading (huge stutter when use TV sensors maverick or TGP, looking to the horizon). Persian Gulf at least don´t stutters with TV/TGP zoom cameras looking at horizon. There must something in caucasus that downs performance related with how terrain is loaded in 2.5.6 I was hoping ED had resolved those issues thorough today´s patch. Its the same. I don´t undersand how VR area could take this hit in performance, most modules are unplayable in VR, specially on Caucasus.
  9. I have the same performance as previous 2.5.6 iterations. Caucasus its unplayable, it sames something related with high CPU usage and terrain loading (huge stutter when use TV sensors maverick or TGP, looking to the horizon). Persian Gulf at least don´t stutters with TV/TGP zoom cameras looking at horizon. There must something in caucasus that downs performance related with how terrain is loaded in 2.5.6 I was hoping ED had resolved those issues thorough today´s patch. Its the same. I don´t undersand how VR area could take this hit in performance, most modules are unplayable in VR, specially on Caucasus.
  10. The same performance as previous 2.5.6 iterations. Caucasus its unplayable, it sames something related with high CPU usage and terrain loading (huge stutter when use TV sensors maverick or TGP, looking to the horizon).
  11. Many Thanks BigNewy for the news, I hope to have at least the same performance we had with 2.5.5.
  12. Thanks, It could be possible to post the changelog, please?
  13. Hi, in other less dense areas, stutter doesn´t happen. However, in a dense area, the stutter phenomenom is non existant with others AC. I attach a zip showing similar high AOA maneouvers, comparing 2 tracks: F-18 (smooth without stutteing) and F-16, in the same scenario. Something happens with CPU utilization with F-16, it seems that high AOA increase a lot the CPU usage (in comparison with other A/C like F-18, A-10, F-14, F-15...). EDIT -- In the F-16 track, its specially notiaceable just 4 seconds before crashing into ground, when inverted. version tested: 2.5.5.41962, in VR (2.5.6 is completely unplatable on my PC - high end - 7700K 5.1 Ghz, 2080ti, SSD M2, 32md DDR4 3200 mhz, Valve index). Sincerely, I hope to expect performance optimizations in near future, at least with the same level of 2.5.5, not worse. Thanks F-16 High AOA Stutter_2.5.5.41962.zip
  14. Only reflects your own chopper. Not exterior...
  15. Here is the tracks. With 3 differents planes, in all of them the gear light is bugged (viewed with valve index hmd), it produces headaches... with 2.5.5, those glitches didn´t happened. F-14-A10C-F15C 3D DOUBLE VISION GEAR LIGHT.zip
  16. Devs this is "checking" since more than 4 months. I attached the necesary info even a trk. Please.
  17. Thanks Bignewy, but... C´mon... nobody in DEV Moscow team have a HMD unit just to test ALL glitches in VR?? my god. There is no one or two issues, there are MANY of them... double vission, clouds, objects disappearing, cockpit illumination, light-landing projections (It hurts my eyes when you´re landing at night, and see everytime the double cutted lights...) ...VR development would be a priority in the sim. Honestly, it seems to me nearly abandoned area by all TEAM (even in VR forum sections, no item is ticked or followed, - it seems to me- nowadays). I´m not speaking about improving,... if only not breaking more things since 2.5.5 I would be happy. Its vey disappointing to me the treatment in this area. 2.5.6 its a totally disaster in VR simulation.
  18. Its failing, all anti-radar missiles in my case don´t hit targets (failing by 20 meter more or less). 2.5.6 latest update.
  19. Just remember the extreme stuttering that occours with tv/ir cameras usage. Its totally unplayable now.
  20. I would expect a better explanation in the newsletter about core sim features current state, seeing such huge performance downgrades in 2.5.6 worries me alot (about -50% specially on VR). Instead of showing resources spent on infantry 3d modelling..., IMHO this has zero interest in a combat flight simulator. Taking into account that maybe those high details models will cap fps performance more than now... Instead doing those things, please, focus now all effort in improving / fixing performance, vulkan support, new/improved graphic engine, etc
  21. Why this is not ever investigated by ED?
  22. Its not a bad idea, just for enhance to 90 fps maybe... you get inmersion surrounding your view, BUT without deep perception.
  23. What surprise to me a lot is that the changelog of today´s patch, ... nearly 100% of changes are totally cosmetic (appart aim-120...), in manuals, in missions, and this is the best... "adding a button to not disable sound when app minimizes" what is that? is that really necessary, and why spend resources on that? More than one month passed since 2.5.6 disaster launch... just to add only cosmetic changes or mission correction? c´mon
  24. Maybe missed adding a minus sign just beforte the "50%", just to reflect at least the beta reality and 2.5.6 disaster in performance test.
×
×
  • Create New...