Re: [Flightgear-devel] WS3.0 progress

[Alex B. <ale...@gm...>]

Hi Salvatore:



> I noticed a lot of stuttering and frame rate which with regular pattern
> was going from 30/40ms to almost 1second.
> I switched off WS3.0 and the situation is better, but still I can see
> stutters.
>
> ...with the same rig I fly regularly on the F-16 in DCS in complex and
> populated tactical scenarios and I have no stutter and fluid and constant
> framerate according to the detail selected, in my case with middle high
> visual setting DCS holds 50fps very stable.
> So moving to FG yesterday and WS3.0 was making me feel uncomfortable
> during the flight. And I was not expecting this, considering DCS a
> benchmark for my rig.
>

It's my understanding that this behaviour is exactly what most people are
seeing of FG these days.
The FGUK Saturday evening multiplayer flights are normally  filled
with people getting the random stutters and causing issues when trying any
sort of formation flying (especially when it's the leader)

It's been how FG has been for a lot of users for several years at least I'd
say.
Not so much a problem when doing generic GA or Airliner operations but it
can be an issue if you're flying something with higher performance or to
tighter tolerances.



On Mon, 23 Jan 2023 at 07:54, Salvatore De Astis <
sal...@gm...> wrote:

> Hi,
> yesterday Iàve spent some time flying WS.0. The scenario was the one of
> all Italy generated from @merspieler.
> Setup:
> - FG build on my windows desktop from latest NEXT source code
> - CPU Intel i7-4790K overclocked to 4.6GHz
> - GPU NVIDIA RTX 2070 Super
> - 32GB Ram
> - FG and scenario installed on SSD drive
> - I used the F-16
> - I use headtracker addon with OpenTrack
>
> I noticed a lot of stuttering and frame rate which with regular pattern
> was going from 30/40ms to almost 1second.
> I switched off WS3.0 and the situation is better, but still I can see
> stutters.
>
> ...with the same rig I fly regularly on the F-16 in DCS in complex and
> populated tactical scenarios and I have no stutter and fluid and constant
> framerate according to the detail selected, in my case with middle high
> visual setting DCS holds 50fps very stable.
> So moving to FG yesterday and WS3.0 was making me feel uncomfortable
> during the flight. And I was not expecting this, considering DCS a
> benchmark for my rig.
>
>
>
> Il giorno dom 22 gen 2023 alle ore 23:09 Stuart Buchanan <
> stu...@gm...> ha scritto:
>
>> On Tue, Jan 17, 2023 at 7:55 AM Erik Hofman wrote:
>>
>>> On 1/16/23 16:51, David Megginson wrote:
>>> > I believe so. Gdb gives me a SIGKILL, with no opportunity for
>>> backtrace.
>>>
>>> It's very well possible this it the new OOM (Out Of Memory) killer in
>>> action.
>>>
>>
>> That seems most likely.
>>
>> After some valuable consultancy from James and Richard, I've created some
>> scenery using PNG rather than DDS for the landclass texture.  Rick ran it
>> this evening and was able to fly for far longer around KSFO than before,
>> without continual memory usage increase.
>>
>> It's available here:
>> https://drive.google.com/file/d/1j91qImcU28jrt290O5sUHbRZe7-fQJf-/view?usp=share_link
>>
>> I'd be interested in people's experiences and whether it improves memory
>> behaviour.
>>
>> (I've also added it to the wiki page:
>> https://wiki.flightgear.org/World_Scenery_3.0 )
>>
>> My theory is that the DDS scenery I've created is borked in some way for
>> some people - possibly as the nvvt library is very fragile.
>>
>> For the use-case we have (granular small textures where we do not want
>> any interpolation or compression), PNG is a better choice.
>>
>> -Stuart
>>
>>
>>
>>
>>
>> _______________________________________________
>> Flightgear-devel mailing list
>> Fli...@li...
>> https://lists.sourceforge.net/lists/listinfo/flightgear-devel
>>
>
>
> --
> Salvatore De Astis
> _______________________________________________
> Flightgear-devel mailing list
> Fli...@li...
> https://lists.sourceforge.net/lists/listinfo/flightgear-devel
>