XP 11.51b1 - still experiencing nasty stutters (with 1080Ti, 8700k and 32GB ram).

Load up Zibo or Default 737 at JFK 31L (max objects). Set view 1 cockpit looking forward. View 2 looking down towards center pedestal. Now press 1, 2, 1, 2, 1, 2, 1, 2, 1, 2.  Severe stutters are present.  This is a test case - but proves there are coding issues to fix (VRAM related).

"XP12 is years away".  Yup sure is. Vulkan blogging started in March 2016. Vulkan "final" release 11.51 or 52 will be approx 5 years later.  Maybe longer.  This is a graphics port.  Nothing new. Therefore, the time to code a whole new XP release would be at least 3 years from now. And that's a best case scenario!

Ok, Greazer.  Let's put this to bed.  You've been asked to post a log file...several times.  You never have.  I'm giving you the opportunity to post a log file so I can take a look at it.  No one can help you if you don't post one.  I can almost guarantee you, it's 1 or more plugins causing your problem.

XP12...years away.  Hey, you're entitled to your opinion on the rather crude timeline you've posted.  But there are those of us who have a direct line of communication to Laminar.  So, whatever you want to believe, feel free to do so.  Even if someone from Laminar came out and told you EXACTLY where they are up to, you wouldn't believe it anyway.  So there's no point arguing that with you.  

The proverbial ball is in your court.  

27 minutes ago, Greazer said:

XP 11.51b1 - still experiencing nasty stutters (with 1080Ti, 8700k and 32GB ram).

Load up Zibo or Default 737 at JFK 31L (max objects). Set view 1 cockpit looking forward. View 2 looking down towards center pedestal. Now press 1, 2, 1, 2, 1, 2, 1, 2, 1, 2.  Severe stutters are present.  This is a test case - but proves there are coding issues to fix (VRAM related).

"XP12 is years away".  Yup sure is. Vulkan blogging started in March 2016. Vulkan "final" release 11.51 or 52 will be approx 5 years later.  Maybe longer.  This is a graphics port.  Nothing new. Therefore, the time to code a whole new XP release would be at least 3 years from now. And that's a best case scenario!

Ye still running xplane on your 286 with 2mb ram and 64meg voodo card but blame everyone else but yourself, I think your here to poke sticks no more.

FYI, unacceptable for XP12 major release:  bad performance, billboard clouds, broken shadows, crumby weather system, forever loading, buggy ATC, murky horizon, limited view distance, floating mountains, overblown haze, single core app!

6 minutes ago, Greazer said:

FYI, unacceptable for XP12 major release:  bad performance, billboard clouds, broken shadows, crumby weather system, forever loading, buggy ATC, murky horizon, limited view distance, floating mountains, overblown haze, single core app!

log file, Greazer.  Let's see what's giving you problems.  If you don't post it, I'll take that as a sign that you're lying about every problem you're experiencing.  

Actually, this doesn't look like "severe stutters" to me.  Max settings.  Looks pretty smooth, actually.

(Greazers latest video)
 

 

Edited November 15, 20205 yr by GoranM

55 minutes ago, Greazer said:

Load up Zibo or Default 737 at JFK 31L (max objects). Set view 1 cockpit looking forward. View 2 looking down towards center pedestal. Now press 1, 2, 1, 2, 1, 2, 1, 2, 1, 2.  Severe stutters are present.  This is a test case - but proves there are coding issues to fix (VRAM related).

All settings pretty much maxed, overcast enhanced cloudscapes. 55-60fps (55 on pressing 1, 60 pressing 2)

[email protected], 32GB RAM, GTX1070 LINUX (hpc-compute power profile) 

I generally get in the 40s, lower power profile and doing other stuff at the same time. There's a fair few nasty cpu hogs still around during cruise but cockpit views isnt one of them. Your problems are elsewhere.

Edited November 15, 20205 yr by mSparks

Oh no hes cast his line that smell y bait has us hooked, im going down, help. plghh, oh thank goodness spat out the bait. Actually @GoranM I think hes writing his own flight simm, lets see  how good he is. As I said before didnt realize how blessed I was till he came along.

16 hours ago, mSparks said:

At least that is something else Laminar is finally not being so quiet about:

https://developer.x-plane.com/2020/11/arm-macs/

Yes, 1050Ti is slower than the M1 integrated graphics chip:

https://www.tomshardware.com/news/apple-silicon-m1-graphics-performance

Further stated in the article in X-Plane:

"... that Apple doesn’t just have fast chips for their new machines, they might have the fastest ones."

The fastest ones, yes sure thing. It`s like someone is ringing on your door-bell. It`s also not an Apple chip, but a licensed custom made ARM architecture. Someone is wearing deep black sunglasses.

 

Better be prepared for bad news in X-Plane? I really would have hoped X-Plane would have been solely developed in the United States.

 

2 hours ago, BigDee said:

It`s also not an Apple chip, but a licensed custom made ARM architecture.

There are four CPU instruction sets (aka "architecture") to choose from, i86 (Intel) i86_x64 (jointly owned by AMD and Intel), ARM (owned by ARM holdings) and RISCV (open source, very similar to ARM).

There are some others like MIPS, but they are "old" RISCV.

Its an Apple chip because they own the physical design for the chip, ARM basically sell the schematics. There is a lot of push away from i86_x64, not just from Apple, the chips are demonstrably power hungry, have stagnated for the last 15 years (Intel is still on 14nm from 2014, the new Apple Silicon is 5nm) and massively bloated with things no one wants plus PATRIOT act backdoors (in mobiles they are in the baseband processor).

Edited November 15, 20205 yr by mSparks

3 hours ago, BigDee said:

I really would have hoped X-Plane would have been solely developed in the United States.

Yes, but then it would be only half as good 😁

I know that ARM sells their design and makes them customizable for their customer needs.

It`s ok for the fans and writers that are happy about the new M1 chip, but I was not expecting such a statement from a serious developer. Just look at three of his statements:

-"The take-away here is that Apple doesn’t just have fast chips for their new machines, they might have the fastest ones."

-"Looks like Apple’s faster"

-"I can say geekbench CPU benchmarks match our overall gut feeling of how well X-Plane does in terms of CPU-bound stuff."

So, even in the emulated version, X-Plane matches his overall feeling, M1 might be the fastest one, faster than AMD and Intel.

It`s a matter of days or weeks, to really see if the M1 chip is faster in X-Plane, even in the emulated version.
If it is not, then there is some pretty bad hype going on, I will follow it.

11 hours ago, GoranM said:

Actually, this doesn't look like "severe stutters" to me.  Max settings.  Looks pretty smooth, actually.

(Greazers latest video)
 

It's not always present and you cannot notice from that video, but by doing the specific test case I listed with max objects at JFK 🙄 

12 hours ago, Greazer said:

Load up Zibo or Default 737 at JFK 31L (max objects). Set view 1 cockpit looking forward. View 2 looking down towards center pedestal. Now press 1, 2, 1, 2, 1, 2, 1, 2, 1, 2.  Severe stutters are present.  This is a test case - but proves there are coding issues to fix (VRAM related).

That test case is akin to maxing out the load on your car and then complain about lack of agility in the cornering parcours.

Actually Sidney Just, the guy that coded the transition to Vulkan in X-Plane, explained pretty well what the problem is with folks that still see stutters in the last blog post:

https://developer.x-plane.com/2020/11/stuff-we-are-working-on/#comment-39042

The "normal people" translation is:

"If you run too high detailed scenery, we can either make it blurry or stutter. We chose stutter. But we provide these two command line switches to try at your own risk, so you can get rid of the stutter, but you might see blurry scenery for a second until everything is loaded up when panning around".

Cheers, Jan

 

4 hours ago, BigDee said:

but I was not expecting such a statement from a serious developer

Step ahead of both of us (and quite a few others crying foul).

In the comments he explains why, - power consumption. These are mobile chips, the target is low power consumption, Apple are getting more compute out of 10 Watts than AMD and Intel are getting out of 150Watts.

Apples desktop versions, that consume ~100-150 Watts, are still to come.......

Theoretically they can be 10 to 15 times faster than Apple just put on the market.

Edited November 15, 20205 yr by mSparks