Jump to content

JB3DG

Commercial Member
  • Content Count

    693
  • Donations

    $0.00 
  • Joined

  • Last visited

Community Reputation

225 Excellent

About JB3DG

  • Rank
    Member
  • Birthday 01/26/1993

Profile Information

  • Gender
    Male
  • Location
    South Africa

Flight Sim Profile

  • Commercial Member
    Yes
  • Online Flight Organization Membership
    VATSIM
  • Virtual Airlines
    Yes

Recent Profile Visitors

2,790 profile views
  1. There's a reason I have enjoyed working for him for more than 11 years now 🙂
  2. You aren't. Some of our projects have been lessons learned that we are in no hurry to repeat and we have always tried to do our best to get things to work for users.
  3. Curious which aircraft these were? I know the Phantoms at least had more bugs but they represented a much earlier development style that we wanted to get out of and improve (it took me 4 years of work on them which it should have taken less than 4 months). Our T-38C on the other hand is vastly better and enjoys the favor of AETC's IPs and our upcoming training products are of the same quality.
  4. Its not compatible with MSFS, except for the outer wrapper we use to retain some commonality within the limits provided by the MSFS low level graphics API which is already a repackaging of NanoVG. Our library is based on external low level custom shader based APIs. And MSFS doesn't give us access to that internal rendering pipeline, which is the whole point of my gripe, that we can do stuff in P3D gauge graphics wise that simply cannot be done, or cannot be done to the same performance level in MSFS.
  5. I do have a gripe with the limitations we are getting with the display drawing as they won't provide us access to the internal D3D rendering pipeline like P3D does. I wrote the hardware accelerated 2D vector and raster graphics library in use internally at Milviz and it is leaps and bounds ahead of NanoVG both in performance and ease of use. On the one hand, knowing by experience how hard of a task it is to create a 2D graphics library from scratch I have some sympathy for those like NVG's creators who undertake the endeavor, but what I have seen both with NVG and Direct2D is API designers overcomplicating the task considerably and not taking advantage of hardware capabilities to the degree that they could. There are some displays, particularly on the newer Garmin, Proline, and Honeywell big screen displays that get far too complicated for GDI and NanoVG to handle without turning performance into a slideshow. Custom shaders and draw call batching become critical to simulating these massive and complex displays to the degree we would like to do.
  6. Everyone on holiday at the moment.
  7. As I recall, SpaceX offered 30% or some such figure over the market value to most of the residents of the village who didn't want to deal with the notices to leave their residences due to potential overpressure events, or let them stay if they were willing to put up with it. Boca Chica is home to several spaceflight enthusiasts who have not sold out to SpaceX, and are only too delighted to be up so close to the making of history. Many of them run very active youtube and twitter accounts with cameras all over the site, to such an extent that Elon himself finds their live feeds useful to keep tabs on how things are going when he isn't present and has publicly stated as such on his personal twitter.
  8. JB3DG

    Tornado Alley

    Just managed to dodge getting twisters here in Chattanooga. It was strangely warm for December just before this line hit us.
  9. Most of us who talked about the limitations of the SDK back in the day were talking about what the SDK was capable of at that time. There are still some limitations that we are waiting to have addressed (PMDG is in the same boat), but for the most part many have been addressed and work is moving forward.
  10. Known Colin for more than 10 years (yikes, I'm aging) now and blunt and direct is his usual tone. No malice or arrogance intended, the only thing that raises his hackles is misinformation and direct personal attacks, as well as useless criticisms (eg "its a broke piece of scheiss" vs "XYZ should be doing ABC not DEF").
  11. Its not correct at all. Tilt 0 only displays returns below the clouds at ranges of >= 40nm, is blank at 20, but displays returns at all ranges when above the clouds which is when it should be blank. Hence why I say it is fundamentally broken.
  12. Ok so update: I have the update more or less complete, but I am reluctant to release it at this time because the way the P3D API is simulating tilt and beam geometry is fundamentally flawed (0 returns with 0 tilt at certain ranges below clouds while showing full returns with 0 tilt above the clouds). We are working with LM to get this fixed, although we have no timeline on when this will happen.
  13. Just a side note on the technical aspect (since I'm the only guy on the team who works on the WX and I am swamped with milsim stuff), the latest P3D includes a new weather radar API which I am looking at using to replace the previous rendering system which is heavily dependent on hooks which is why new versions keep breaking it. However, for reasons explained previously, it is not number one on the priority list although it is by no means at the bottom (Curse of 10+ years development experience == high demand on milsim stuff. Gee have I been doing this for that long?).
  14. This is absolutely comical. Carenado has their market base that they appeal to, more power to them. But comparing what they do with what teams like Milviz, PMDG, A2A, MJC etc do, and what's involved in updating their products is just hysterical. I don't know what all the WT team is doing, what information they have on the CJ4's PL-21, and what all you view as a fully simulated PL-21. All I can say after going through the real PL-21 manuals for 3 different aircraft (P-180 Avanti PL-21 manual: 470 pages. KA350i manual: 666 pages. LJ60 manual: 666 pages.) is that it is waaaaaaaay more than just "vectors legs, proper discontinuities, intercept legs, fully functional FADEC, full VNAV, RNAV with all the right sensitivities, including angular sensitivity for LPV, custom LNAV" etc. Again, I can't go into details without breaking NDAs. Suffice to say, there are certain things that I know for a fact cannot be done with the present tools that PMDG, MJC, Milviz, and a bunch of others consider essential for porting over their work into the new sim.
×
×
  • Create New...