Sgt_Baker

Hi Sungsam, This is one that has been asked before, and will be supported by a key/joystick/mouse-combination of your own choosing in a release in the near future. Additionally, under such circumstances, I can also remove the FPS-hit while all displays are hidden. While this is technically possible, there are a number of reasons (technical and "keeping the complexity sane") why I've not included it. That said, with the ever-delayed advent of DCS 2.0 (which would allow me to remove UltraMFCD's FPS-hit in one go), I'm beginning to entertain the notion of including some highly-sophisticated techniques for completely removing the FPS-hit for 1.x.x.x - however this is because those techniques are of interest to my other fields of work. Hope that makes sense. :) Baker Edit: The reason it's been so long coming is that I need to make the keyboard/joystick/mouse code a lot more sophisticated - and that's not exactly the top priority. :)

UltraMFCD 1.0.0.71 is available for download at https://ultramfcd.com. Greetings folks, After something of an election-related hiatus, UltraMFCD development is back to its normal pace. This release contains a whole bunch of technical fixes under the hood, but principally a feature that users have been requesting for ages: A-10C MFCD, CMSC, RWR and Clock can now be resized freely by holding shift while dragging the resize handle. (Similar support for CDU etc. to follow shortly) Best regards, Baker

Hi deadzone, You're not alone in experiencing this error (see my latest post in the other uMFCD thread). Will be working on a solution with immediate effect. Regards, Baker

hi generalcuz, It would appear that you and a chap over in the other thread are experiencing precisely the same issue, going by the diag packages. I'll look in to it immediately. Baker

You'll be able to change the aspect ratios freely in the next release by holding shift while dragging the resize handle - already coded up. :)

I do appreciate what you're saying here. It is not for me to comment further than the technical difference between Fullscreen/Windowed. :) That said, I recently responded to a PM from the 4K crowd regarding the subject of fullscreen support. The official fe23 line is that implementing complete fullscreen support for DCS 1.x.x.x would nearly certainly involve a complete and costly re-write when DCS 2.0 arrives. It was initially my intention to develop both modes side-by-side, but due to the forces of R&D and beta testing, we established much more in the way of functionality using windowed-only. Thus, forces of demand being what they are, the fullscreen effort was resigned to being a border case given the majority weighting. That's not the end of it, however. It is essential for uMFCD to support fullscreen mode in the future given technologies such as Oculus Rift (true 3D being a speciality of fe23 dating back to appx. 2000), TrackIR and so on and so forth. Without committing to any particular date: It will be implemented when it is sensible to do so. :) Baker

We know. The guy in the picture is Rade Serbedzija, aka "Boris the Blade" in "Snatch". He was the person pasted in to a MiG cockpit. The person we were referring after that to was Mower's avatar, being Sean Connery.

:) Absolutely. Has very much stood the test of time. We digress, folks.

Shctating the obvious, Mower, ish the last... ONE PING ONLY, VASHILY!

There was a chap who added a very convincing photo of this actor to one of the MiG cockpits. Have looked for the thread, but can't find it. Anybody else remember it?

What do you mean by "personalise", tovivan?

Hey folks, It would be interesting to gather a little more data on this fullscreen vs. windowed thing. For reasons I'm about to explain, there shouldn't, in this day and age, be much if any difference between the two performance-wise. Firstly, what is the difference between the two modes? Without getting excessively technical, there's surprisingly little difference. All the drawing of shapes, management of textures, lighting, shaders and whatnot are identical. The only difference is what happens when the graphics subsystem (DirectX in our case) wants to finally draw all of that stuff to the screen. In fullscreen mode, the draw occurs to the display's backbuffer, which is a section of memory that represents each pixel on the screen. Once all the data is written, a bit of hardware on the graphics card sends that data out of a port, down a cable to your monitor. In windowed mode, there's simply one extra step. In this instance, the display's backbuffer is owned by Windows, and each DirectX/OpenGL/whatever is drawing to a "virtual" screen. When the virtual backbuffer is ready to be displayed, it is copied within the graphics card's memory to the main windows-owned backbuffer and then it's shoved down the cable as before. So the extra copy operation is why my FPS is dropping, right? Wrong. This type of memory-to-memory copy operation is something that graphics cards have been designed to perform at the hardware level for years (called a "Blit"). Doing some in-my-head calculations, it would be fair to say that an average graphics card could blit a 4K image at 1000FPS without even breaking a sweat. Why does my FPS drop when using UltraMFCD? This is an actual limitation with the way DCS is coded at the moment - specifically the version of DirectX it uses. In order to get the displays out of DCS, I have absolutely no choice other than to copy the data back to system memory. On my system, this takes 3ms per frame of the exported displays (and is why uMFCD has the option to skip frames for the exports). I did perform extensive experiments in order to hack my way around this limitation, but it rendered DCS sufficiently unstable that I would have spent 24/7 chasing reports of mysterious bugs and rendering problems. So, if you suspect Windowed mode is your FPS hit, try running DCS in Windowed mode without UltraMFCD. As it stands, UltraMFCD will reduce the game's FPS by a small amount, although it is entirely dependent on your system config, display settings etc. HOWEVER! It is clear that DCS 2.0 will be using a more modern version of DirectX. This will enable UltraMFCD to completely bypass the FPS hit we experience at present, while maintaining all the functionality we have now, in addition to anything we have planned for the future. So... let's hold our breaths! :)

No. The whole idea is that they remain fixed regardless of where you're looking. :)

Good to hear. Internally we're at about 1.0.0.70 by this point. Getting properly bogged down with the legal shenanigans required to get to full release status. Turns out the lawyers insist that my personal details be attached to this product whether I like it or not. So: My name is Christopher Norman and I trade as fe23. I make cool software when I'm not making things virtually explode. :)

Working now?

Yes, entirely my fault. Will update tomorrow. Sorry, sorry, sorry. :) Edit: Scrap that. I will turn back the time server. :D

Still working on this end, Rivvern. What are the symptoms of the failure? If there's nothing obvious, could you generate a diagnostics package and post it here, please? Details of how to do so are in my sig.

Help is always appreciated. As mentioned, as it stands it is "fitted for but not with" full 3D support. If you're wondering, I used to work in an R&D lab making solid-state (no moving parts), moving zone (move your head around freely) autostereoscopic (no-glasses 3D) display prototypes for the likes of NATO, Samsung etc, along with all the software to drive the above. In other words, real 3D gets me a bit excited. :)

Funny you should mention this. As it so happens we've already developed a technique whereby one can export four full-size/full-colour displays using no more bandwidth than is presently occupied by the A-10C with all exports enabled. Whether all of the F-18C's displays are rendered in colour is up to the developer, so I can't guarantee anything prior to seeing the bird fly... but there are, how shall we say, some tricks? :)

This is something that hasn't been tried yet. I have long-standing experience with stereoscopic stuff, so coded with a view to enabling OR support at some point in the future. Suppose I ought to order one of their dev kits, eh?

In Utero?

This lot might be a good place to start. The issue you're likely to face is that you're entering the specialist monitor market, so costs will likely be much higher than they would be when purchasing consumer-grade kit. Baker

<shameless plug> Or just try UltraMFCD :) </shameless plug>

Heads up: Have just sent a build with free-scaling (hold shift while resizing) of displays to Icemaker, whom at this very moment is demoing his pit at a games exhibition in France.

It really depends on the problem. In my experience, most problems can be made to run in parallel in some way. That is not to say that I deny that certain problems MUST run completely synchronously - thus rendering any multithreading moot. Yup. Not advocating ED suddenly re-write their engine.