Jump to content


  • Content Count

  • Joined

  • Last visited

  • Days Won


Krycztij last won the day on February 5 2018

Krycztij had the most liked content!

Community Reputation

3 Neutral

About Krycztij

  • Rank
    Brig. General
  • Birthday 09/05/1981

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Sounds awesome – record a video!
  2. Even less from my side This Blender automation stuff amazes me. That’s exactly what we need … I wish I could contribute more …
  3. Exactly! 👈 In the meantime, I found out why it’s all so dark: The PSX has a special gamma curve; you probably have to take the square root of each color value to make it look right (i.e. 50 % brightness in the files should actually be ~70 %). I’ll provide an updated file set one day …
  4. No, nothing besides CONTROL=1 should be required. There could be many causes, so I’ll have to repeat the Wine tests myself. TAW uses Wine’s DirectPlay emulation while TFXplorer uses its RawInput emulation, so there’s no connection. Oh and we need a dialog for calibration and control binding, I guess. Don’t get me started. Short version: TAW computes physics in 40 Hz steps. TFXplorer goes up to 240 Hz. The formulas in the F-22 flight physics have some weird force peaks when you fly tight curves at high speed and release the stick immediately (happens often with keyboard controls) and these only surface at high physics computing rates like TFXplorer’s. Then for a very short time, acceleration climbs to 20–200 g and the plane decides it’s too much and explodes. Fixing the F-22 physics is PITA, so I guess I’ll just smooth out the keyboard controls.
  5. This maybe fixed now – Wine 4.0 release notes (https://www.winehq.org/announce/4.0) say
  6. The Internet Archive has a book on Ocean Software, publisher of many games including DID’s EF2000 and Epic: https://archive.org/details/The_History_of_Ocean_Software_Retro_Fusion_Books/page/n263 There’s a few mentions of F-29 Retaliator (pg. 68); ADF/EF2000 (pg. 127), Inferno/RoboCop 3 (pg. 160). Interesting to read about F-29, of which a dozen PC versions had to be written to account for the different types of hardware …
  7. Krycztij


    Looks like ClearType failing on the font. You can try right-clicking this text block in your browser, open the developer tools, and try assigning the standard font "sans-serif" (without quotes) to the element. (Depends on the browser, but I know you’re clever and you can use search engines ) If "ll" comes out better, then it’s a problem with the forum’s Roboto font. (Why ll is special: Some browsers combine certain letters to ligatures because they look better, but some fonts have poor ligature support.) Fun fact: on my Linux machine, all capital Ls in captions render wrong, on all websites, no matter the browser.
  8. No, it does require some 32-bit OCX modules. I‘ll see into that later.
  9. I had to use a 32-bit system for it, but it works. Thanks a lot!
  10. The different places are normal because parts of the CRT have moved from the compiler to the OS since Windows 8.
  11. The reason I didn’t answer is that I don’t know what to do either. nmake configure script hell with unresolved externals is exactly the reason why I never compile any projects from the internet. basic_ofstream is the wrapper for standard output streams like stdout and stderr. If the linker cannot find them, then I’m pretty sure it’s because of the Visual C++ Runtime Library. If not in the IDE, it is controlled via these switches but the documentation must be the worst I’ve ever seen, so let me take an educated guess here: Either the module itself or one of the dependent library projects looks for basic_ofstream but doesn’t find it. basic_ofstream is in the Visual C++ Runtime (CRT). If your program looks for this symbol at link time, that means that somewhere, something is linked statically to the CRT (/MT or /MTd switch). If the linker does not find this symbol, that means that you are linking dynamically (/MD or /MDd switch). You are mixing up static and dynamic linking in one of your dependencies. This is a pretty hard error and I’m afraid I don’t know any easy way to solve it. I would start looking for the /MT or /MTd switch in all dependent projects and, if found, check for a version of the project with dynamic linking instead (which is the default for new projects). Wait, no! The error says it can’t find the __declspec(dllimport) version of the function. So you are probably already compiling against the dynamic CRT, but missing its import library. Check if you are linking libvcruntime.lib and libucrt.lib (or, in Debug builds, libvcruntimed.lib and libucrtd.lib). Visual C++ does link those by default, but it opts out under some rare circumstances (like custom entry points or the /nodefaultlibrary switch). Sorry for your experiencing the darkest side of C/C++.
  12. Yes, they have worked hard towards standard compliance and finally reached it in the first half of 2018. There are some switches for backwards compatibility, but honestly it is far better to delete the old MS-specific cr*p You’re doing great!
  13. Have you tried starting it from the „Developer Command Prompt for VS 2017“ in the start menu?
  14. Yes. The DirectX SDK and the Platform SDK have been merged into the Windows SDK. This is installed with Visual Studio, and you find it under C:\Program Files (x86)\Windows Kits\<version number>. The version number is just important for the feature set and does not define a requirement for the final executable – if you choose the 10.0 SDK but don’t call Windows 10/8/7-specific APIs, it will run on Windows Vista as well.
  • Create New...