X-Plane 11 pb15 is out

[Sesquashtoo 2,846]

4 hours ago, merf said:

XP11. pb15 - For my setup FPS=21 to 23.  so far pb13 was best fps at 25-26.  hoping the final ver hits around 30 would be good.

System: i7 3.4ghz, 16gb ram, GTX 1070

I'm getting exactly the same...21-23 with most sliders maxed.  If xEnviro 1.6 puts out less clouds, can rise to 27-40.

[Paraffin 2,197]

26 minutes ago, N1125Y said:

..sorry, but what's wrong with a Rig that is running at 4.6GHZ and a nice, considerably, modern GTX970, do we need  something like the supercomputer "TaihuLight" to run this Sims ???????

No, it depends on your settings and your monitor resolution.

The X-Plane developers have always tried to stay a little ahead of the available consumer hardware, so there is room to grow over a cycle of 4+ years between major versions. The idea is that you shouldn't be able to max out every setting, and still run at the highest possible monitor resolution (4K, 8K, VR?). By the time it's towards the end of the version cycle, the hardware will catch up.

If you don't have a supercomputer, but you do have an average "AAA gaming" computer setup, then that should be enough.... after the beta is finished, because it's still being tweaked. Just knock down the settings enough to get the frame rate you need.

For reference, I'm running a i7 6700 4.0 GHz, 32 Gig RAM, GTX 970 video setup. The new pb15 is knocking down a few frames from what I had before the new clouds came online, but I'm still getting between 37 and 48 fps over most scenery. That's with Graphics set to High/HDR and maximum world objects. The critical factor there, I suspect, is that I'm only using a 1920x1200 monitor. Anyone pushing a 4k monitor won't get that frame rate, and will need more hardware grunt. People who post about low frame rates too often don't mention what monitor res they're running, and there are some crazy-high res monitors being used out there now.

Anyway, gripes about frame rates at different settings and with different hardware will mean more, once XP11 goes final. We're still in the framerate optimization stage, and they just introduced a new cloud model with a major impact.

[N1125Y 22]

.hopefully they will come with final in my life time ...... anyway, with the last B15 I went down to stable 3-5FPS (yes, 3-5FPS) from 42-50 prior to the update .... and so help me G_d I did nothing just updating, reboot the Rig and fire up the Sim ..... I think from here I can go only up ..so I predict that by the time RL will be out with  B35-B40 update, I'll be back to 45FPS ..lol 

My CPU is i5-4690 at 4.2GHZ and the GPU is GTX970 16Gig of DDR3 at 1866Mhz 2 SDD HD ....

 

One happy simmer

       

[Jimmy RFR 406]

Installed, booted up, played with weather for half an hour, did a quick circuit, shut it down and spent the weekend involved with other projects.

However, from what I can see performance was about the same, just with a bit nicer clouds. No stutters. I think under some conditions the clouds could use more tuning - they pack pretty small and dense in heavier coverage.

Nothing seemed broken for me, so that's a plus.

[RJC68 823]

11 hours ago, N1125Y said:

.hopefully they will come with final in my life time ...... anyway, with the last B15 I went down to stable 3-5FPS (yes, 3-5FPS) from 42-50 prior to the update .... and so help me G_d I did nothing just updating, reboot the Rig and fire up the Sim ..... I think from here I can go only up ..so I predict that by the time RL will be out with  B35-B40 update, I'll be back to 45FPS ..lol 

My CPU is i5-4690 at 4.2GHZ and the GPU is GTX970 16Gig of DDR3 at 1866Mhz 2 SDD HD ....

 

One happy simmer

       

Try deleting everything in your preferences folder. I do this every time an update comes through. I make sure I keep a copy of the joystick & key preferences so I do not have to re assign all my controllers

[Mountain Man 142]

18 hours ago, N1125Y said:

.. this Sim drive me nuts ..run the latest update restarted the Rig, fire up the Sim and I'm down to 3-5FPS (yes, 3-5FPS) ...... cross my heart I did nothing but updating ... is this what RL meant Optimization ???

Bottom line, this thing is not flyable ... I guess I'll tow it back to the hangar, get some Popcorn & cup of coffee at the Executive-Terminal (KSPS) and wait  ... unless of course you guys have any other ideas .....lol

In all probability they haven't started any serious optimizing yet.  That won't happen until just before release because no programmer wants to put time and effort into optimizing "live" code that could very well get rewritten or discarded during the next phase of the project.

I think some people here have the wrong idea about software development.  Here's how it doesn't work: You cobble together an alpha build that is incomplete and full of bugs, and throughout the beta process, there is steady and predictable improvement from one version to the next as the software is carefully polished to a state of perfection.  No, really, get that idea out of your head.  Software development can be a messy, unpredictable process.

[Gdavis0101 0]

I am still getting the pause and go problem that was introduced with Threaded Optimization being turned ON, it is turned off Globally but after installing PB 14 this problem is back, PB 15 did nothing to rectify it. In between pauses the sim runs fine and it ran fine before PB 14. Submitted bug report, will hope for the best.

[rototom 198]

21 minutes ago, Mountain Man said:

In all probability they haven't started any serious optimizing yet.  That won't happen until just before release because no programmer wants to put time and effort into optimizing "live" code that could very well get rewritten or discarded during the next phase of the project.

I think some people here have the wrong idea about software development.  Here's how it doesn't work: You cobble together an alpha build that is incomplete and full of bugs, and throughout the beta process, there is steady and predictable improvement from one version to the next as the software is carefully polished to a state of perfection.  No, really, get that idea out of your head.  Software development can be a messy, unpredictable process.

I don't know if I follow what your saying. It is obvious that they would try to optimize performance, however, the part about programmers writing some type of un-optimized code, or temporary code in case it's dropped? In our shop, when software goes to any stage of testing, the code is expected to be completed to design spec; we test what we plan to deliver. It will need to change in order to address defects and performance issues, but developers do not intentionally promote some type of intermediate code that is un-optimized expecting to change that later. Once you've address all of the defects and issues, how can you go back in and start changing code? You tested the code, and now your changing it? I realize that it's a cycle of change/fix and re-test etc..but why would you test code that you know isn't going to be in the final product?

 

[Longranger 148]

3 minutes ago, rototom said:

I realize that it's a cycle of change/fix and re-test etc..but why would you test code that you know isn't going to be in the final product?

Yes and no. In a lot of cases you can´t solve a problem in one go. Instead you have to establish a new infrastructure that will only make sense at a later date.

But I don't expect a huge amount of performance 9increases before X-Plane 11.0 goes final. Most improvements will now target the new clouds. While they work quite good in a lot 0of cases, some people seem to have real problems. I woulöd think the new clouds could become the death sentence for GPUs with only 2 GB of VRAM.

[N1125Y 22]

6 minutes ago, rototom said:

I don't know if I follow what your saying. It is obvious that they would try to optimize performance, however, the part about programmers writing some type of un-optimized code, or temporary code in case it's dropped? In our shop, when software goes to any stage of testing, the code is expected to be completed to design spec; we test what we plan to deliver. It will need to change in order to address defects and performance issues, but developers do not intentionally promote some type of intermediate code that is un-optimized expecting to change that later. Once you've address all of the defects and issues, how can you go back in and start changing code? You tested the code, and now your changing it? I realize that it's a cycle of change/fix and re-test etc..but why would you test code that you know isn't going to be in the final product?

 

+1 with rototom .... I understand that in our world launching a B version is for the purpose of getting feedback and final tuneup. however in XP11 case I think that LR release a product that is not Beta But more Delta Version, it is way to Pre-Mature, & IMHO the current situation will increase resentments. After all I'm a customer not a hamster .... 

My sim is not flyable and nobody can help .... (-:        

[wiler 394]

2 hours ago, N1125Y said:

+1 with rototom .... I understand that in our world launching a B version is for the purpose of getting feedback and final tuneup. however in XP11 case I think that LR release a product that is not Beta But more Delta Version, it is way to Pre-Mature, & IMHO the current situation will increase resentments. After all I'm a customer not a hamster .... 

My sim is not flyable and nobody can help .... (-:        

totally disagree, its completly usable. there are bugs but its not that far off. the fact that your system may not be up to task is a different story....

[sd_flyer 6,038]

PB 15 seems like best update ever for me. I got record max fps running x-enviro with fps hungry JD A330

[Richard Sennett 2,499]

25 minutes ago, sd_flyer said:

PB 15 seems like best update ever for me. I got record max fps running x-enviro with fps hungry JD A330

Is the Jar A320 & A330 working ok with PB15 ? thanks

[Mountain Man 142]

3 hours ago, rototom said:

I don't know if I follow what your saying. It is obvious that they would try to optimize performance, however, the part about programmers writing some type of un-optimized code, or temporary code in case it's dropped? In our shop, when software goes to any stage of testing, the code is expected to be completed to design spec; we test what we plan to deliver. It will need to change in order to address defects and performance issues, but developers do not intentionally promote some type of intermediate code that is un-optimized expecting to change that later. Once you've address all of the defects and issues, how can you go back in and start changing code? You tested the code, and now your changing it? I realize that it's a cycle of change/fix and re-test etc..but why would you test code that you know isn't going to be in the final product?

First you get it working, then you optimize it for performance.  My point is that software development is not a linear process.  Sometimes a solution to a problem is non-optimal, or it causes problems somewhere else, and sometimes you have to throw out an idea and try something else.  Yes, some optimizing will happen during development, but the bulk of it won't be done until just before release.

Just hold your horses and let Laminar do their thing.

[Mountain Man 142]

3 hours ago, N1125Y said:

+1 with rototom .... I understand that in our world launching a B version is for the purpose of getting feedback and final tuneup. however in XP11 case I think that LR release a product that is not Beta But more Delta Version, it is way to Pre-Mature, & IMHO the current situation will increase resentments. After all I'm a customer not a hamster .... 

My sim is not flyable and nobody can help .... (-:        

You bought a beta, dude, and Laminar was completely upfront about what you were paying for.  Stick with XP10 for now.