48 minutes ago, Moria15 said:

From what I can see so far through my own personal testing is that on my specs, there is no loss of performance on DX12, but if there is any gain it is a couple of frames if that, and it could be just local conditions, which even if set identically, are subject to minor change.

What I have seen, by using AMD hardware is that the use of multi-threaded mesh shaders, which are new to DX12 seem to me (and this is personal) to provide a slightly better depth of colour and density.

In addition, it is noticable that my CPU utilisation has dropped to an increase in GPU usage, and my memory footprint has lowered slightly and my core utilisation on the CPU showes more usage of the multi cores.

Considering that this implementation is literally just replacing the wrappers and allowing the hardware to utilise it's built in functions, I hold out hopes for some real improvements when optimised with real DX12 code, especially on modern hardware.

I am using the latest AMD GOTY drivers which I think were optimised for the new DX12 workflow and DX12 multi-tasking shaders.

Graham

Good to see... i have also updated everything but haven't tried dx12 yet. Will do so tonight since i have some time off to evaluate.

On 11/20/2021 at 10:39 PM, Sesquashtoo said:

1.  Water (ocean, lakes, rivers, streams) colors totally different depending on the depth and angle of the sun

2. Sky colors...more intense and not washed out, even in haze conditions.

3. Terrain...the best rendering so far of natural color, (shaders!) and even better running MSFS-G. I see the difference clearly between the shaders of DX11 and DX 12 Beta.

I also picked up in most places an extra 1 or 2 FPS, but nothing below my normal 29/30 floor.

 

Right after GOTY update, this was exactly my feeling after trying dx12, the colors and shades were better, I saw it right away. And yes fps was about 1 fps higher (I was doing a careful controlled test).

But every 30 seconds or so if I am not just on AP, there were stutters. Also if changing view direction very fast for several seconds there was stutters and fps drop. Which did not happen in dx11.

The annoyances finally sent me back to dx 11 permanently.

6 minutes ago, Fielder said:

But every 30 seconds or so if I am not just on AP, there were stutters. Also if changing view direction very fast for several seconds there was stutters and fps drop. Which did not happen in dx11...

But I am getting this in DX11 also now after this update. 

Before, I could go to external view and circle around and around the aircraft forever, and it was silky smooth.  Now it seems to skip  / stutter / pause every 360 degrees. 

I am not sure if it is coincidental, but even at vastly different locations, it seems to skip as it faces north.  I can't pin it down to any scenery in the direction of the pause.  All very strange. Still investigating.

35 minutes ago, Fielder said:

ut every 30 seconds or so if I am not just on AP, there were stutters. Also if changing view direction very fast for several seconds there was stutters and fps drop. Which did not happen in dx11.

Like I mentioned earlier, the frame timing in DX12 is awful 

On 11/21/2021 at 5:35 PM, Sethos said:

The same forum full of people that loves anecdotal evidence, "I think, I feel, I see" and not a single 1:1 comparison shot despite being on page 3 now, for something you could easily validate with a screenshot. 

Wow I feel at home here! Sounds like the speaker cable / snake oil debate in hi-end forums!

I know how this is going to end .... fetching popcorns ....

A.

27 minutes ago, bobcat999 said:

But I am getting this in DX11 also now after this update. 

Before, I could go to external view and circle around and around the aircraft forever, and it was silky smooth.  Now it seems to skip  / stutter / pause every 360 degrees. 

I am not sure if it is coincidental, but even at vastly different locations, it seems to skip as it faces north.  I can't pin it down to any scenery in the direction of the pause.  All very strange. Still investigating.

This may not help at all, but I had to set Offscreen Terrain Precashing to Ultra to get rid of stutters when spinning the view around. It surprised me that Ultra was way more stutter free than lower precashing settings.

On 11/21/2021 at 11:35 AM, Sethos said:

The same forum full of people that loves anecdotal evidence, "I think, I feel, I see" and not a single 1:1 comparison shot despite being on page 3 now, for something you could easily validate with a screenshot. 

https://share.icloud.com/photos/0JuGG_MeWeFqe3XLmbcq18xtw

5 hours ago, ErnstB said:

Thank you!

I am still on 456.71 with good performance, but will change to your 471.68 one.

Ernst

As we all know...there truly isn't a fire-burner 'improvement' between each set...but, I have had such good solid top FPS and non stutter performance with 471.68,  I have just stayed with it....why change?  The actual/any benefits of constantly updating to my view, is so small, that as they say;  "if it ain't broke DON'T FIX IT!"   LOL.   Let me know how you feel the suite works for your GTX1070. That would be interesting, and cheers!

Edited November 22, 20214 yr by Sesquashtoo

1 hour ago, Fielder said:

Right after GOTY update, this was exactly my feeling after trying dx12, the colors and shades were better, I saw it right away. And yes fps was about 1 fps higher (I was doing a careful controlled test).

But every 30 seconds or so if I am not just on AP, there were stutters. Also if changing view direction very fast for several seconds there was stutters and fps drop. Which did not happen in dx11.

The annoyances finally sent me back to dx 11 permanently.

My experience with using DX12 on my four core i7 Extreme 980/GTX1070 has not seen that, so I guess I am thankful for that, in being that I really am enjoying whatever 'changes' to date I am seeing in now running MSFS-G under the DX12 'wrapper'.  Much, much (for myself) nicer ground texture depth and colors, tree foliage colors, and I am really liking what it has done for all water bodies, where shadows come into play like at the coastlines....no more artifacts...good shading, so for me, a keeper. Maybe as we see on this thread, for other's a pass...each will choose.

Edited November 22, 20214 yr by Sesquashtoo

1 hour ago, ADamiani said:

Wow I feel at home here! Sounds like the speaker cable / snake oil debate in hi-end forums!

I know how this is going to end .... fetching popcorns ....

A.

I don't think this will end up locked...people will 'impart' their take...and move on.  Some will as I have, see changes, subtle or not,...some will not...will state that...and as they say, '"that's that", lol.  All good...no popcorn needed for the thread 'long haul'. LOLOLOL! Besides, it gets in your teeth....

3 hours ago, Ianrivaldosmith said:

Like I mentioned earlier, the frame timing in DX12 is awful 

absolutely. and since a picture and video says more than a thousand words:

while DX12 may provide about the same fps as DX11 (but often less than DX11) it shows consistently higher frame times which creates those stutters. up to 246 milliseconds is a quarter of a second! to achieve those 20 fps in the top right screenshot the frame time would have to be as low as 50 milliseconds, 5 times lower. in other words: looking at the fps only can be very misleading. a constant frame time of 246 milliseconds would only provide 4 fps! the reason why MSFS even with DX11 runs pretty smooth even at conservative low fps is because of its pretty constant frame times, see lowest line graph on the left, in contrast to constant frame time spikes on the right side with DX12. just numbers, no opinion 😀

Edited November 22, 20214 yr by turbomax

4 hours ago, Ianrivaldosmith said:

No one ever talks about frame time. In DX12, the frame time is all over the place. 

you are right. frame time is a little more complex subject than fps, that's why no one talks about it.

its FRAME TIME, yay! 

 

Edited November 22, 20214 yr by turbomax

To me DX12 seems to have a bit more contrast and saturation, I prefer the look of DX12.

I expected a horror show with DX12, but it looked better contrast and colors.

20 hours ago, Sesquashtoo said:

I don't think this will end up locked...people will 'impart' their take...and move on.  Some will as I have, see changes, subtle or not,...some will not...will state that...and as they say, '"that's that", lol.  All good...no popcorn needed for the thread 'long haul'. LOLOLOL! Besides, it gets in your teeth....

Well maybe not locked but you know, when it's about audio cables I am in the field of those who actually hear the difference, so I know the non believers can be pretty nasty :-)))))))

I always say: to each its own.

And I actually saw a small difference in color saturation. Then I went back to DX11 because the stutters were all over the place, but yes I saw something there.

A.