Ubisoft engineer says Microsoft/Sony pressuring them to force 30fps on PC.

DX12 is not primarily about Mantle-like efficiencies, it doesn't matter when development was started, these things can be put into the development process.

If Mantle is work in progress, what is DX12?

You will see that development of Mantle is suspended in the near future. Mantle isn't being put into DX12, it's being copied with or without AMD's direct help.

 

I don't know what you mean by Mantle like efficiency, but DX12 is all about efficiency, it's why Intel have been all over it.

Obviously DX12 isn't finished, I was just stating that neither was Mantle.

Mantle isn't close to being copied, if it was AMD's lawyers would be involved, the fact unlike Mantle, DX12 works on a wide range of old and new hardware suggests the two are very different.

 

I thought M$ said that DX12s main focus was unification, making console ports easier?

AMDs lawyers wouldn't be involved if AMD are participating. Just watch this space and see how Mantle disappears in the next year.

This article pretty much sums up my feelings

http://www.pcworld.com/article/2109...-pc-gamings-software-supercharged-future.html

 

I don't think I've ever read MS claiming that DX12 was there to make console ports easier.
Personally i don't see why they would want that anyway, unless it's a MS game it is of no financial benefit to make other developer's job easier.
Not that making ports easier is a bad thing, it's actually a very good thing.

I do expect Mantle to disappear, but only because AMD will have a hard time promoting it.

That's an old article and it's speculation that I don't really agree with, if anything it makes more sense that MS developed DX12 for products like the Surface Pro as high efficiency and low power usage is what these products rely on.

 

Because MS made a big hardware investment. So of course they are going to have some exclusives to that hardware. MS does release PC games though, which wouldn't happen if what you say is true.

 

Just read that reddit note on motion:

http://i.imgur.com/ofURqEh.jpg?1

I think that's correct, although not properly contextualised

 

Where's that from?

Even those who hate high framerate movies agree that it's smoother, they just think it's looks horrible.

If there was no difference, there would be no fans or haters.

 

It's referring to motion, and essentially blur.

If I run an after effects composition or C4d or w/e, and crank the motionblur samples upto 24 times (per frame), you would not be able to tell me there is a difference above that.

I believe that is what that is referring to in video games, not the actual quantity of frames per second.

The picture is from a reddit user straight off of a textbook.

 

I think with movies more frames is also better (just as with games), but it does have a clear downside as compared to games. "Defects", Mistakes, ... in special effects and what not show up more easily with a higher framerate.

But for games there's no discussion imo. Smoothness, input lag, ... so many things improve massively with higher framerates and personally I think 120/144 fps is about the limit where we can't notice much improvement (at least not with current hardware, taking curring delays of input devices into account).

 

It's all down to what you're used to, I only have experience with interpolated movies and low budget footage, and both look cheap and low budget.
I've also heard it makes big budget movies look cheap and horrible, but is great for those who get eyestrain with 3D.

When it comes to games, nearly all PC gamers are used to 60fps, so 30fps looks bad and/or painful.

 

There is absolutely no way this is correct. 24FPS are not enough for most objects to make a transition pixel-by-pixel across the display. Faster objects moving across the display require more FPS in order to appear in perfect motion. The eye does not work in FPS... if you were to control a very slow object, a few FPS are enough, but if you were to control a very fast object, hundreds of FPS will be welcome.

This is independent of motion blur. If you were to strobe x ms per frame at any framerate, higher framerates would still offer more information and would appear smoother and more whole, although you're only getting x ms of motion blur (which again varies in number of pixels based on how fast the object is moving).

Try this experiment (I wish you had a 120Hz monitor): Lower your mouse sensitivity and move the mouse in small circles. Raise your mouse sensitivity then do the same. You will notice the lower sensitivity appears smoother, since the cursors are drawn closer apart than at a high sensitivity. This is particularly apparent at higher refresh rates such as 120Hz.

Fast-paced FPS games benefit enormously from the increased motion detail. Strategy games become smoother to pan and motion blur is reduced. Racing games become much more fluid, even if you were only watching them.

Try a 120Hz monitor. Don't game on it. Just watch the game run. See the difference.

No amount of artifically added motion blur will make up for less FPS. The solution is to have high enough FPS and let our eyes and brain take care of the motion blur. The real world does not blur in front of us.

 

Tell you what - I'll upload two pictures, one with 24 motion samples per frame and one with 128 (maximum I can get) and you will not be able to tell me which is which.

I agree that 30 frames per second is pretty poor, but motion blur (I believe) is where the apparent confusion lies on this.

 

I just explained that factoring out motion blur by strobing the image would still allow higher FPS to be smoother. There is no wasted motion information if the object in motion is only in apparent motion, skipping across the display a bunch of pixels every time.

What do you mean by uploading 'pictures'?

Of course I'll be able to tell the difference.

 

https://flic.kr/p/pMvxgF this is picture number one

https://flic.kr/p/pv5dDt this is picture number two

Both pictures are identical compositions taken at identical moments in the timeline.

I used 24 frames per second, 90 degree shutter angle for motion blur - a standard film configuration.

One of them is running at 24 motion samples per frame and one of them is running 128 motion samples per frame, both to induce blur. The composition itself is a simple rotation of the word GURU3D, rotating 720 degrees horizontally over a 1 second period. The picture is taken exactly halfway at the 12th frame.

To help you out, this picture https://flic.kr/p/pMA1yW is only using 3 samples per frame, we'll call it picture number 3.

So, which is which?

 

What good is a still picture? We're discussing motion.

I don't understand.

 

The pictures show motion, for which I believe is accurate based on what the reddit user's picture of the textbook describing the ability to not detect motion samples greater than 24 is based off.

If you like, I could upload the 2 versions of the 1 second video for you to look at, but that will take time to do.

 

That would be great.

 

file number one

http://s000.tinyupload.com/index.php?file_id=75767260792446418095

file number two

http://s000.tinyupload.com/index.php?file_id=84224970643765446124

I used H265 for the best quality so you should have a pretty clean file to look at.

 

I still don't understand what you're trying to do.

I just downloaded two files that are both running at 24 frames per second.

How in the world can I see a difference between different FPS if both videos are at the same framerate?

 

Chalk me up to also not getting the point of this. Increased frame rate helps make things more immersive and better captures fast moving images. The poster seems to be confusing in another concept entirely.