Performance problems with the A-model

[Wali763]

Hi.

 

Im on monitor/TrackIR.

 

In our squad we have some training maps, one for Syria. When spawning on the Theodor Roosevelt (SC-module) on that map with the A-model all is very stuttery on the carrier, meaning, that I have usually ~55fps but with big spikes in the frametimes, which makes the image stop every few seconds for a blink. Plz see pic. Looking at a E-2 taking off in front of me, it stops for a few times running down the catapult. Also the deckcrew is kind of brake-dancing....

 

After the E-2 and S-3 have taken off, Im alone on the deck and still when aligning for the catapult, the deckcrew is doing its brakedacing... terrible!

 

With the Hornet or even the B-model, this does not happen and I usually have the 60fps (I capped it at for TrackIR) and only tiny spikes (which do nothing regarding stutter).

 

I also had three recent crashes (Windows or only DCS freezing) on that map and Im still not sure, if this is realted to DCS/F-14A or a bad OC. About the latter I tested a few things and increades Vcore, but still had one crash since then. Also went to CPU-stock or even to 5,2Ghz from my standard 5,0Ghz. Stock did seem to make it even more shaky but 5,2Ghz did not help at all.

 

I have a tracklog, but hat is 5,01MB, so too big for an upload. But I can easily provide another one, if needed.

 

Once in the air, its not a problem any more.

 

In general I rarely have witnessed stutters on DCS with my settings, thats why im reporting this. Its not a too big problem for me, since Im usually prefer other modules, that have not problems, but since the A-tomcat came out just this week, I spent quite some time getting used to it.

[fat creason]

Very unlikely that it's purely the F-14A causing this, what are you system specs and settings? Are you flying in VR?

[Wali763]

Hi.

 

All this is rather unscientific, I know. Im on monitor/TrackIR.

 

Plz see the pic. I did this a few times and the results are moderately consistent, but I would not call them conclusive. In the same scene, the Hornet had almost no stutters (=> spikes) at least none I would witness, but the F-14A had more than the F-14B.

 

In tests done earlier, the difference between F-14A and F-14B were more obvious then in this "round" of tests. Still I can say, that I almost never experienced stutters with the older B-model and this only came to my attention when testing the newer A-model.

 

Nevertheless, I found that exchanging the SC-Theodore Roosevelt with the non-SC John Stennis resolved the whole thing almost completly. I kind of feel, the deckcrew could be causing this, since often such spikes coincide with them doing something like giving signs etc.

 

So my guess is, that the Supercarrier (known to be CPU-hungry) in combination with the Tomcat (also known to be a bit more CPU-hungry compared to eg the Hornet) might cause this. So my plan is to avoid these situations (on this big Trainingmap) for the forseeable future. I just used this one anyway to do some familarisation with the A-Model and to do some stresstesting to my system since as mentioned above some crashes happend on this map.

 

Still, from doing this mission-start about 40 times now, I saw the A-Model being consistently more stuttery than the B-Model.

[fat creason]

What's your computer hardware? CPU/GPU/RAM?

[Wali763]

After rethinking and retesting another mission (SyriaChemStrike, where I exchanged the B for an A and did not see a real difference) Im not sure, if my case is not too specific (maybe only on that special Trainingmap??) to justify a bug-report.

 

My system is a i5-9600k, oc-ed to 5,0Ghz, 32GB of 3600/CL16 RAM, RTX 2070Super, Samsung 970Evo 512 GB M.2, 1440p 144Hz-monitor. Usually (when not capping for TrackIR I have up to 120fps higher up, and down to ~45fps low on Syria-Map. I had it in my bottom-line but after the forum-update thats unfortunately gone and I have not yet taken the time to look into this.

 

Anyway, I will investigate this further and report back if I find something conclusive.

 

If you say, getting a 9900K would solve the issue, Id surely do. They arent that expensive (330Euros) anymore right now. Actually much better value than the overpriced 5xxxX from AMD right now. :smilewink:

 

Thx for your help/input anyway and sorry for the stress.

[fat creason]

Your system is significantly better than mine and I don't think I have performance issues like that. 99% of my DCS time is SP flight testing on the Caucasus map though, so I'm not running any complicated missions.

[Wizard_03]

FWIW I'm seeing the same inconsistences as well, some days mission is buttery smooth, sometimes same mission is a slide show. I have a very similar set up to the OP. I can't pin point the problem, tried the meta/fxo delete tried repairing.

 

All my other aircraft are rock steady AFAIK haven't tried the F-14B though

[Jeepyb]

+1 came to report the same thing. F14A butter smooth on Caucasus but I get a stutter every 20sec or so (seems to be regular interval) on the Syria Map. Tested with the F14A quick action case 1 supercarrier missions for both caucasus and syria.

[Biggus]

I'm seeing similar seemingly random changes in performance from day to day, but with both Tomcats. I'm suspicious that this is an ED problem rather than an HB one.

[Wizard_03]

I've started to jump into a mission then quickly exit and jump back in. At least for my custom missions that seems to help. Haven't played with the stock ones.

 

I think could be something to do with mission initialization. It seems like when I get these dips they are associated with system changes like a new RWR spike or Mode 4 datalink info on the TID poping in, Radar mode change, HUD mode change. ect. ect. But after I do my restart trick usually everything is groovy. So I definitely think it is an  F-14 issue. It happens for me on maps like Nevada in the middle of the desert at high altitude or out over the water on the PG map.

[Wali763]

Hi.

 

Just went back and retested my situation with the new patch. And - at least for  now - it seems to have gone. After the intial loading straight 60fps with no stutters at all!

 

Great!

Edited December 18, 2020 by Wali763

[fat creason]

That's nice to hear. We didn't change anything on our end at least.