fixed Performance (FPS) loss from 2.7.6 to 2.7.7

[Raven434th]

4 hours ago, Mr. Big.Biggs said:

Continue to monitor but haven’t flown or purchased anything for almost a year now. Just can’t keep up with all the configuring every time they release a patch or update. Tired of the slideshow. 
That said, here’s the thing that chaws me off a tad. Every update we keep seeing a portion of folks that report fps drops. 1.5% here, 3% there but it adds up. For the flat screen guys, they can absorb this and still enjoy the experience. AND that’s GREAT!
 For the VR guys, it’s just not worth it anymore. Yes I choose vr and yes I prefer MP so I get it, that’s on me and that’s not my gripe. 
My gripe is they continue to pile load after load of fps sucking goodness on a program that worked awesome when I bought it, but has now been rendered pretty much useless for my needs(again I said my needs so save the fanboy slams) and the icing on the cake is Bignewy continues to come on here and say “sorry guys, plays great on my end, just not seeing it” with a machine common to many of us and  using the current generation of VR headset.

So he either fibs to tow the party line, or knows something he’s not telling….. 

end of rant  have a great weekend  

 

 

 

 

 

Gotta admit I thought the same thing...kinda can't help but think it ...but I also have a nagging feeling that this isn't the case because I have a hard time thinking Wag's wouldn't see these threads as well and make sure it got addressed. I'm HOPEING it's a "gonna give em' a great Christmas present" kinda thing and they don't wanna spoil it..........I'll be pretty pissed though if come Christmas I'm still running below 30 FPS. my wallet will close if that happens. Going from DCS at 20-30 fps and looking horrible to 60 plus fps and looking beautiful  in iL-2 is taking it's toll.

Edited November 20, 2021 by Raven434th

[Boosterdog]

35 minutes ago, Mr. Big.Biggs said:

Naturally if I would have posted without that delineation I would have been bombarded by the “my flat screen works awesome” crowd so I choose to speak on my behalf. Not for a second diminutive to your cause so please don’t shift focus off of the POINT  of the argument onto the punctuation….

Not here you wouldnt. And it wasnt me shifting the focus off the point and onto VR 

[Mr. Big.Biggs]

We appear to be on the same team. How about acting like it. Sorry to bruise your fragile sensitivities… 

[jumphigh]

On 11/18/2021 at 10:09 PM, deep said:

 If we keep 2.7.6, can we have all the current aircraft fixes by downloading the modules again? Or do we have to have a current DCS World to match?

Like others said you won't get any module update. You can stay on the older version not more than one year because after that you will loose authorization and you have to update if you want to play again. I wanted to stay on 2.5.5 with my old RX570 4GB but this March I ran out of authorization and was forced to upgrade. No way to stay on the older version. DCS might be a moving target but they can't expect that their customers are always able to upgrade hardware if they raise the toll. So without LTS versions my pocket remains closed – I do not buy stupid DRM-crippled SW any more.

With my old RX570 card I suffered FPS losses from 2.5.5 to 2.5.6. I got over 60 FPS with 2.5.5. That went down to between 50 and 60 FPS on 2.6. With the 2.7-track I managed to get 46 FPS average. But now with this "stable" version 2.7.7 it won't go higher than 40 FPS. Every revision measured in UH-1 Free Flight mission.

I'm playing in 2560x1440 and all settings Low/Medium or off. 4GB VRAM isn't a problem, it stays below 3,7GB on the Caucasus map. But even with going deeper to low textures, shadows off and tree and theatre density to 0.1 DCS is capped below 40 FPS!? Everything is looking bad, worse than games from the late 90ies. Ridiculous! CPU (i5 10400) stays at 10% and RAM is below 12GB. What the hell are they doing???

Edited November 20, 2021 by jumphigh

[Limaro]

Dear everyone,

as you may have notices, the thread to the performance loss has been closed, due to endless discussions. I opend this one, as ED asked that everyone having issues, should open his own thread, posting as much data as possible to support the ED developer team.

Even though the thread being set to "reported" as a ED tester could have reproduced it, I asked the question, if it is still whished, that everyone having the problem should share his experiances in a own thread - and this was answerd true.

I truely believe, that this has only advantages:

- First of all, there would be a reliable number of affected people. Espacially because data will be a bit better prepaired.
- And it offers an easy way fo ED to "collect" the benchmars, dxdiags,.. (nobody wants to crawl through the pages of a thread).
- It would also give ED the possibility to give individual feedback.

So I kindly ask everyone of you, to not overload this thread, as I tried to share data here, that is ment to help ED finding a solution. And If i find something new, I'd like that it does not get lost.

@impalor:  You said you did not feal strong performance difference from 2.7.6 to 2.7.7. That is very interesing for me. Do you have the possibility to install the 2.7.6 again (maybe using stable or OB, which one you dont use) and make a real benchmark? And if there is really no difference, please share then you dxdiag and settings.That would really make me curious

[jumphigh]

They should make a automatic benchmark including log with the free Frogfoot on the Caucasus and Marianas maps. That will deliver standardised and repeatable measurements to the devs. Requesting tracks, FPS and frame times from customers is a big mess...

Edited November 20, 2021 by jumphigh

[Nevyn]

6 hours ago, Mr. Big.Biggs said:

 For the flat screen guys, they can absorb this and still enjoy the experience. AND that’s GREAT!

If only.

[Raven434th]

19 minutes ago, Nevyn said:

If only.

yup...22inch samsung here

[Miro]

6 hours ago, Mr. Big.Biggs said:

 For the flat screen guys, they can absorb this and still enjoy the experience. AND that’s GREAT!

 

 

 

 

 

Well NO, I'm on flatscreen, my woman is on flatscreen, we got very different hardware, and both got slideshow. We thought we'd come back to the Caucasus, but it's also bad there. We were waiting for AH-64, but if we can't fly Hind, the more we won't fly AH-64, it's good that we didn't buy it, because it would stand and wait for a miracle...

[Mr. Big.Biggs]

Hey, I will claim ignorance on the flat scree thing.
 Other than the magnificent cinematography of the grim reaper’s videos tho only thing I have to go on is the dozen's of “you VR guys are the problem, all great with my track IR” replies.
If its a mess across the board so be it. Didn’t intend to offend… 

[Boosterdog]

3 hours ago, Mr. Big.Biggs said:

Hey, I will claim ignorance on the flat scree thing.
 Other than the magnificent cinematography of the grim reaper’s videos tho only thing I have to go on is the dozen's of “you VR guys are the problem, all great with my track IR” replies.
If its a mess across the board so be it. Didn’t intend to offend… 

Olive branch - yup- your right. Im an over sensitive little pixie. But I will raise you a "TIR is great" for a  " VR is god"   

We are on the same team in this matter. Lets do as Limaro says and make a new thread with the data to prove the point. If nothing else it avoids the issue being wrote off as the problem of a few.  S!!

[TAW_Impalor]

13 hours ago, Limaro said:

@impalor:  You said you did not feal strong performance difference from 2.7.6 to 2.7.7. That is very interesing for me. Do you have the possibility to install the 2.7.6 again (maybe using stable or OB, which one you dont use) and make a real benchmark? And if there is really no difference, please share then you dxdiag and settings.That would really make me curious 

Will try, curious myself

[scampaboy]

19 hours ago, Boosterdog said:

No we cant. AND its not GREAT!  Making out its just a VR thing is plain wrong and also dimishes the argument for taking action by reenforcing a MYTH that it affects only a realtive small proportion of higher demand players. I see this type of comment in virtually every performance thread including the many actually started by a flat screener. No doubt a VR capable rig will play the game in 2D incredibly. But not everyone can choose to have such a rig. Not everyone can afford the £1000 GPU minimum upgrade  to bring the horsepower now required to replace even 10 fps.  As a result they find themselves up exactly the same creek as Gavin in the Goggles.   

Those goggles may expand the virtual world but they appear to induce a certain level of tunnel vision also...

A comment I must sadly admit that I took with a gain of salt. 

So did I, it actually annoyed me as they have been promising performance improvements for the past few months. I fly 2d and and I’m making baby steps into VR. 
Any improvement I fear will be swallowed up in “features” that should have no affect on performance. 
I feel for people new to DCS who spent money on a computer before the last update and then lost 10% performance for no benefit. It must be soul destroying. 

15 hours ago, jumphigh said:

They should make a automatic benchmark including log with the free Frogfoot on the Caucasus and Marianas maps. That will deliver standardised and repeatable measurements to the devs. Requesting tracks, FPS and frame times from customers is a big mess...

 

They wouldn’t sell any modules then when people see how their system really performs 

[scampaboy]

15 hours ago, Limaro said:

Dear everyone,

as you may have notices, the thread to the performance loss has been closed, due to endless discussions. I opend this one, as ED asked that everyone having issues, should open his own thread, posting as much data as possible to support the ED developer team.

Even though the thread being set to "reported" as a ED tester could have reproduced it, I asked the question, if it is still whished, that everyone having the problem should share his experiances in a own thread - and this was answerd true.

I truely believe, that this has only advantages:

- First of all, there would be a reliable number of affected people. Espacially because data will be a bit better prepaired.
- And it offers an easy way fo ED to "collect" the benchmars, dxdiags,.. (nobody wants to crawl through the pages of a thread).
- It would also give ED the possibility to give individual feedback.

So I kindly ask everyone of you, to not overload this thread, as I tried to share data here, that is ment to help ED finding a solution. And If i find something new, I'd like that it does not get lost.

@impalor:  You said you did not feal strong performance difference from 2.7.6 to 2.7.7. That is very interesing for me. Do you have the possibility to install the 2.7.6 again (maybe using stable or OB, which one you dont use) and make a real benchmark? And if there is really no difference, please share then you dxdiag and settings.That would really make me curious

I remember a conversation I had with my mother in law years ago. There was an advert on the notice board of the block of flats advertising a subotnik that weekend in the yard. I said to the mother in law are you going to go and help should I go ? And she said to me “For what should I do someone else’s work for nothing that I have already paid them to do myself”

Maybe they should sort it out themselves as they have been paid and not depend on the community to sort it out. 

[jumphigh]

50 minutes ago, scampaboy said:

They wouldn’t sell any modules then when people see how their system really performs

I don't know. I remember the benchmarks inside of ARMA 2. One of them showed single-digit FPS. Now, 10 years later, decent CPUs and GPUs don't change that much. Nobody trashed the game because of that.

As a low level heli pilot sometimes I think the graphical details of DCS are worse than that old gem. I know the maps are huge and you can't compare to hand crafted maps, but especially towns and villages aren't that sparse and empty in reality. Even the new Marianas are looking empty. So I really don't know why DCS is so slow.

[scampaboy]

6 minutes ago, jumphigh said:

I don't know. I remember the benchmarks inside of ARMA 2. One of them showed single-digit FPS. Now, 10 years later, decent CPUs and GPUs don't change that much. Nobody trashed the game because of that.

As a low level heli pilot sometimes I think the graphical details of DCS are worse than that old gem. I know the maps are huge and you can't compare to hand crafted maps, but especially towns and villages aren't that sparse and empty in reality. Even the new Marianas are looking empty. So I really don't know why DCS is so slow.

Not trashing the game, I have loved DCS for years probably will for many years to come. 
I tried to help a guy on Facebook a few weeks ago who was asking what VR headset or tracking he should buy to go with his system and get decent frames. The poor guy was actually under the assumption that he would be able to run a VR headset on the minimum system specs quoted on the website. 
After following peoples advice to download it try the free aircraft and be careful with his expectations he did and decided to uninstall. 
They either need to sort out the performance problems or change the recommended specs to more reflect reality. 
 

Anyway I’m going to go and fly and manage my expectations

21 minutes ago, jumphigh said:

I don't know. I remember the benchmarks inside of ARMA 2. One of them showed single-digit FPS. Now, 10 years later, decent CPUs and GPUs don't change that much. Nobody trashed the game because of that.

 

They aren’t trying to sell modules on the back of that software though are they 

[TAW_Impalor]

@Limarohmm. Downgrading to 2.7.6 disabled all my modules! Repairing and renaming saved games did not help. At first also lost main menu screen, but repairing fixed that. Used Skatezilla's UI. Any ideas? )

Edited November 21, 2021 by impalor

[jumphigh]

34 minutes ago, scampaboy said:

They aren’t trying to sell modules on the back of that software though are they

Bohemia Interactive tries to sell expansions and DLCs for ARMA 2 and much more for ARMA 3. It isn't that different to DCS. And I remember how I got into DCS: I bought a disc copy of UH-1 in Germany from Koch Media in 2014. I didn't know that DCS even existed that days – the disc copy looked like a stand-alone sim. To my surprise I got a free Frogfoot on that – and never played it. So I wonder how many enthusiast out there are getting into DCS just with the free base module?

I took me several months until I realised that you can fly trough trees. Since then I always thought, WTF, it's a helicopter sim, isn't it? I took years until version 2.0 to change that ridiculousness. There are users inside the UH-1 subforum willing to pay for a new UH-1. I say: Eagle Dynamics, get the job done for the old one! Collision boxes on the ground are another point of ridiculousness.

Every iteration something breaks or is worse than before. I lost all trust in the devs.

[Limaro]

57 minutes ago, impalor said:

@Limarohmm. Downgrading to 2.7.6 disabled all my modules! Repairing and renaming saved games did not help. At first also lost main menu screen, but repairing fixed that. Used Skatezilla's UI. Any ideas? )

 

You mean it does not start any more? Thats why i meant to install the other Version. OB when you normally run Stable, or visa versa.

[TAW_Impalor]

51 minutes ago, Limaro said:

You mean it does not start any more? Thats why i meant to install the other Version. OB when you normally run Stable, or visa versa.

No space on NVMe. Installing on HDD may not be apples to apples, but I'll try.

[Limaro]

14 minutes ago, impalor said:

No space on NVMe. Installing on HDD may not be apples to apples, but I'll try.

That does not matter. I've made most of my tests having 2.7.6 on my HDD, but it was still faster.

[Boosterdog]

2 hours ago, impalor said:

No space on NVMe. Installing on HDD may not be apples to apples, but I'll try.

If you wanted a like for like you only really need the base game for the purpose of a comparison. You dont even need a module. Given you seem to have wiped your modules, its probably a great time to just install Stable 2.7.6 base and OB 2.7.8 base afresh. If you want to be extra careful,  before you download the second install, temporarilty rename the first installed game's root folder (ie DCS World to DCS World-temp). This is ensure common files are not copied from one to another. After the install is finished - revert the naming. 

[TAW_Impalor]

5 hours ago, Boosterdog said:

If you wanted a like for like you only really need the base game for the purpose of a comparison. You dont even need a module. Given you seem to have wiped your modules, its probably a great time to just install Stable 2.7.6 base and OB 2.7.8 base afresh. If you want to be extra careful,  before you download the second install, temporarilty rename the first installed game's root folder (ie DCS World to DCS World-temp). This is ensure common files are not copied from one to another. After the install is finished - revert the naming. 

Repairing OB back to the latest restored my modules. I have no space on NVMe for both, and installing them to HDD is just too much pain. For benchmarking I use F18 8v8 mission because it is the closest to what I normally fly.

Nevertheless, I installed Release version to HDD (took 3 hours!) and benchmarked 2.7.7 vs 2.7.6 build. Result: I saw a drop of about 1 FPS out of 45.

The latest OB on my NVMe drive runs faster, but that is not apples-to-apples. On NVMe my FPS never drops to 0, unlike on HDD. See this link for detailed results and my configs/optimisations/useful scripts.

Nota bene: I will hopefully upgrade to Alder Lake next week and my signature will change. So I included my 'old' PC details too.

[Boosterdog]

11 hours ago, impalor said:

Repairing OB back to the latest restored my modules. I have no space on NVMe for both, and installing them to HDD is just too much pain. For benchmarking I use F18 8v8 mission because it is the closest to what I normally fly.

Nevertheless, I installed Release version to HDD (took 3 hours!) and benchmarked 2.7.7 vs 2.7.6 build. Result: I saw a drop of about 1 FPS out of 45.

The latest OB on my NVMe drive runs faster, but that is not apples-to-apples. On NVMe my FPS never drops to 0, unlike on HDD. See this link for detailed results and my configs/optimisations/useful scripts.

Nota bene: I will hopefully upgrade to Alder Lake next week and my signature will change. So I included my 'old' PC details too.

From what I can see from the data, and my very limited (read zero) understanding of VR, your system is getting cained pre and post 2.7.7/8.  Steam VR is doing its "magic" and reprojecting- a lot- to present even the minimum 1/2 rate of 45 fps to you. Not sure if you change the repro manually or its automatic, I assume its automatic.

If thats correct, its the VR equivalent of me and my flatscreen with vstink running say 55 fps in 2.7.6 and 40 fps in 2.7.7/8.  In both cases the vsync software will change down a gear to half refresh rate (30 fps) because my system cant produce the 60 fps vsync requires just as yours it half stepping to the 45 fps.- as such my perceived performance will not have changed in the same way yours hasnt. I'd be getting 30 fps like i always did. In reality however the loss is there, I just cant see it cos Im already at a low state and I wont see until I cant make 30 fps.  Difference here is that Stream VR has further tricks up it sleeve in the shape of reprojection to help keep things bearable. 

What Im saying is that you cant make a meaningful comparison with fps alone in such circumstances as there there are more factors, external to DCS, at play. As an Analogy, you cant tell a Fiat Punto (2.7.8) has an overall slower top speed than an Ferarri (2.7.6) by driving them both at a speed limit of 45 mph (half rate) with cruise control (Steam VR) on. 

Thats why we switch off vsync when comparing. In your case I think you are in that ballpark where the perfomance difference is "soaked up" by an already choked output. If you pushed up your settings so stream VR algorithms couldnt cope or were previously on the cusp, I'd venture you'd experience a difference. 

Forgive if Im wrong but I suspect your set up is masking the lost performance at this time. As it probably is for many others and likely the reason why "where the F has my performance gone" posts tend to drip into the forums and not tsunami as each player hits that magic mark in their own time.  

Edited November 22, 2021 by Boosterdog

[speed-of-heat]

I always recommend turning off asw/motion smoothing to measure performance for the same reason you recommend turning off vsync