Dx 12, yes or no?

[216th_Jordan]

As the question came up to whether or not the game will be updated to Dx 12 or Dx 11:

How do you stand towards dx 12 and why?

 

Dx 12 has great benefits computing wise and also reduces CPU load but is a Windows 10 feature only.

Dx 12 is compatible with all Dx 11 Nvidia cards and following Ati cards: http://www.guru3d.com/news-story/amd-released-list-of-compatible-directx-12-cards.html

 

 

I'm not totally into that topic as a developer is that is for sure, but that is why I'm hoping for a good discussion about this here!

 

Thank you for your participation

Edited June 19, 2016 by 71st_AH_Jordan

[216th_Lucas_From_Hell]

If they went up to DX 12 I would upgrade my OS and GPU, but a lot of people wouldn't so I see why they could be reluctant about going to DX 12. Most of the regulars (i.e. the people who are here on these forums) would do so too, but the casual players wouldn't bother with that and then you'd lose a lot of the online population and go through a low point in sales as well until enough people have made the transition.

 

For example, according to the Steam Hardware survey from May, Windows 10 is the most popular OS, but at the same time it constitutes only 40% of the players. Another 40% are Windows 7 and 8.1, which can upgrade but there is a catch - the free upgrade period ends on the 29th July, so after that anyone who hasn't upgraded will have to pay for it, potentially reducing the number of conversions. Even during the free upgrade period, W10 has only gotten 1.5% new users since April, meaning the vast majority of W7/8 users are staying where they are for a while. Remember how many people were running Windows XP well into 2013? It's looking that way.

 

A poll here has shown similar results, with 50% on W10, 40% on W7 and only 10% on W8.

 

On a work laptop here with W10, everything is roses and I'm loving the new OS, so no complaints on my end. However, some people have shown dissatisfaction for different reasons.

 

Overall, despite how powerful DX12 is, I don't think it's financially viable to do the upgrade just yet, even if it will bring notable improvements to performance. The main points to balance here are: how popular will VR be in flight simulators (% of players), how much benefit will DX11 bring (enough for the foreseeable future or not), how much benefit can DX12 bring over DX11 (in FPS), how much will the upgrade to DX11 cost vs. upgrading to DX12, and if this would be recouped in a time frame that makes sense. If DX12 brings bombastic improvements but people don't use it because their OS doesn't run it, then there's no point, right? Do keep in mind that a decent share of their market is not in the richer parts of the EU and North America, so while users in these countries can easily cash out for gear that runs it, others won't be able to do so.

[Lusekofte]

I would not make a upgrade to Win 10 just now, not for this game sake. But I am sure I will some time within a year or 2. Because my rig is having problems keeping up. Problems related to such exclusive feature will be loosing customers. 

Win 7 is far too popular at the moment, but soon it will not be supported and more people will migrate to 10. 

I think dx 12 is the way to go, but too soon

[216th_Jordan]

You arguments are valid. given that: would a rewrite to dx 12 be a big deal if we would have dx 11?

[coconut]

You arguments are valid. given that: would a rewrite to dx 12 be a big deal if we would have dx 11?

 

Yes, and that's something people not familiar with graphics programming might not understand, but the benefits of dx12 require significant architectural changes to one's code. It's not just a matter of adapting to slightly different data types and functions. Nothing comes for free. The big benefits of dx12 are easier to achieve with a fresh new engine.

 

I am also not convinced that dx12 will bring big improvements to the game. Sure, it would allow for more detailed terrain, but the level of detail we have now is already pretty good. It would also allow for more trees and grass, but it won't help with physics, AI and network processing.

 

Of course, the step from dx11 to dx12 is smaller than between dx9 and dx11.

 

Ultimately, I'm just guessing and I'd like to believe those are educated guesses, but only the devs can have a (somewhat) precise idea of the costs and benefits. Our opinions on the subject are hardly relevant. Players shouldn't care what version of directX a program uses, only what features the game delivers, and what kind of computer you need to take advantage of that.

[Feathered_IV]

I would rather they improve upon the shallow gameplay first.

[Lusekofte]

I think if they have long term plans for this game they got to migrate to DX 11 or 12 . We are in a middle of a small revolution when it comes to customers expectation. The VR technology is up and coming , Oculus Rift have started to gain momentum and cheaper brands are surfacing. People start buying high end GPU and or migrate to consoles for experiencing VR.

There will be improvements in this way of gameplay, and simulators are a excellent game for VR play.

I hesitate to upgrade my PC for this reason, I think within a year PC will be built for the purpose of VR and one can know for sure just how much money you really need to put into such a Rig

[coconut]

Voted yes to everything, because of course I'd like to have a dx12 version, but see my post above for reservations. Basically I'd love to get a super-modern engine, but I would hate for the studio to go under while trying to build one.

[GrendelsDad]

I am with Coconut...I am still blown away by what the Devs have done with DX9, I can only imagine what they could do with a now well developed DX11.   But I too vote for yes on all.

[Lusekofte]

I look at this game like the Fairly Battle, it was sooooooo modern in 35 and totally obsolete in 39. I think a modernization will be forced in a year or two . Or it will be abandoned

[SharpeXB]

If the team is going to put the work in for a new game engine. The only one that makes sense is DX12 as it won't likely be updated soon. If they feel the market isn't ready. Then just wait till it is.

By the time such a new engine would be ready probably the market would too.

71% of the market already has a DX12 capable GPU*

39% of those people are stil running W7 but most people don't upgrade their OS unless there's a good reason. Well a game running DX12 is a good reason.

And if the future is VR. DX11 isn't adequate for the task like DX12 is. You can read the VR reviews of another sim elsewhere. But DX11 still relies on commands being sent through the CPU and for games like flight sims that render huge numbers of objects DX11 can't generate a consistent acceptable frame rate of 90fps for these headsets.

The game already looks beautiful in DX9

But the next gen engine should aim higher than DX11

 

* From the Steam Hardware Survey. Flight sim players are likely to have stronger hardware than the average person. For example

50% of the people on that survey only have dual core CPUs. They would even be able to run BoS very well at all.

 

I'm sure 1CGS knows the hardware stats on its own buyers and can decide based upon those numbers.

Edited June 24, 2016 by SharpeXB

[Toxin1]

Not just an upgrade but a move to a new engine with DX12 or Vulkan API would open things up for the IL-2:BOx franchise and also likely carry over to a new ROF2

[Jade_Monkey]

Pretty sure Han recently hinted DX10/11. So i dont think DX12 is coming anytime soon.

 

Im on my phone, so i cant find the quote as easily. I'll try from home later.

Also, for those wondering, Win10 is a very good OS despite all the creepy things you can read about privacy.

 

I dont have a single complaint.

[J2_Trupobaw]

Switching exclusively to DX 12 is great way to alienate the customers who are not switching to Windows 10... such as myself. To much bother for one game. DX11 is the safe and sensible option at the moment. 

[Dakpilot]

The benefits in performance and future proofing with DX12 outweigh the "safe'' outdated sensible option

 

there really are no legitimate reasons not to go windows 10 other than 'I don't want to'... the only legitimate reason would have been the cost, but is a non issue because of the free upgrade

 

In a shortish time when DX12 is mainstream and all new games are written for DX12 people will look back and realise that an outdated DX11 was a big mistake, a re-write for DX11/12 would take time and be expensive

 

before you know it BoS/BoM would be left behind again and a great opportunity would have been missed, because it would be a very very long time to wait for another game engine re-write..

 

anyway that's only my opinion, and you know what people say about opinions 

 

Cheers Dakpilot

[SharpeXB]

Having a favorite game available in DX12 would be reason enough for Windows 7 users to upgrade. That's not much of an obstacle.

[216th_Lucas_From_Hell]

Not sure you can make major decisions exclusively based on people who have this as their favourite game. A lot of people are casuals, and the hassle of an upgrade (which after this summer will not be free any more) can easily put those off, which is troubling.

 

Maybe they could ride it out on DX9 until the numbers start looking brighter for DX12, and then undertake such an upgrade.

[L3Pl4K]

With dx12,the game need less cpu power for the same number of draw calls/3d performance.

http://www.eurogamer.net/articles/digitalfoundry-2015-why-directx-12-is-a-gamechanger

http://www.pcgamer.com/what-directx-12-means-for-gamers-and-developers/

So there is  more cpu time for physics, ai units, sound effects.

This test shows, that older gpus would also benefit from dx12. More draw calls compared to dx11.

http://www.anandtech.com/show/9112/exploring-dx12-3dmark-api-overhead-feature-test/3

I think an upgrade(new engine) to dx12 is expensive and need some time. But it is cheaper to pay 50-80 Euro/Dollar for an engine upgrade, as many hundreds for a high end cpu/gpu brut force soulution.

Edited June 25, 2016 by L3Pl4K

[Dakpilot]

Hassle of upgrade..really, one click and that was it , for those who have not done it, and I believe it runs out end of July,...

 

If you do not like it you can revert  to Windows 7 easily, but not to take advantage of free Win 10 now....at least to try?

 

Which people will be the first to bitch about the need to spend $+100 upgrade to Win 10/benefits of DX12 ..after the summer......

 

If a game/engine is written/created for DX12, is it not possible to make it playable/backwards compatible for DX11 users? I understood other games can/will do this?

 

Remember DX11 is 2009 tech and DX9 is from 2002.....

 

Cheers Dakpilot

Edited June 25, 2016 by Dakpilot

[216th_Lucas_From_Hell]

Absolutely Dakpilot, but clearly there is a crowd that would still be stuck in Windows XP if they could, and now are holding for their dear life onto an OS from 2009 because the aliens are presumed to be spying... or something along those lines.

[Dakpilot]

^^ +1..there will always be some who wish to stay with "Windows XP" and am sure they have their 'reasons'...this however should not stop progress....

 

Cheers Dakpilot

[J2_Trupobaw]

 

 

there really are no legitimate reasons not to go windows 10 other than 'I don't want to '... the only legitimate reason would have been the cost, but is a non issue because of the free upgrade

 

 

 

That's a wishful opinion presented as a fact in hope to shout down people who disagree. The legitimate reasons for not switching to Windows 10 yet is

 

1) Untried

2) Untested

3) Years of experience with Microsoft OS show that only one in three is actually worth the switch

4) And the ones worth switching to became such after few yars and obligatory service pack, not in release year

 

in short - to immature a product to isk committing a perfectly working gaming PC to it. Your quote should actually say

 

"there really are no legitimate reasons not to go windows 10 other than 'I don't want to waste time and effort on setting up a new PC from the scratch and working through inevitable issues only to find out the result is inferior to what I had, forcing me to repeat the process only to get where I started, to risky until it becomes proved product.'...

 

"Not broken, don't fix" is a perfect reason for not switching, too.

Edited June 26, 2016 by Trupobaw

[Dakpilot]

More than 300 million people are using Win 10 now

 

By the time a game engine re-write would be completed this number will be exponentially larger (1 billion projected in 2018) and DX11 will be more than 11 years old tech

 

The performance advantages of DX12 are too important (especially with regard to Flight sims) to be ignored

 

In the computer/software world, to heavily invest in 2006 (released in 2008) tech for a new product in 2016/7 does not seem to make financial or logical sense

 

DX11 worked, was not broken and did not need fixing..but still remains very out of date   the (software) world is moving at a faster pace these days, like it or not

 

Cheers Dakpilot

[6./ZG26_5tuka]

Looking at DCS Dx11 will surely be capeable enought to turn this sim engine into a smooth and stable platform. Combine that with the fact that it also easier to implement in short time and more "proven" tech I'd be pretty much satisfied if they leave it by that.

 

I'm with Tupobraw, a single game is not worth all the hassle and issues that come with a fashion upgrade.

Edited June 27, 2016 by 6./ZG26_5tuka

[Dakpilot]

Calling the performance improvements of DX12 a "fashion upgrade" is a bit steep    the massive reduction in CPU load even compared to DX11 is the very thing Digital Nature Engine requires for most of the features called for

 

Only reason DCS update is DX11 is that they started it at least 5 years ago

 

You are all Luddites I say!!     Stay with your trusty 2008 Nokia 1202, I will be embracing what we have available in the 2016 phone market (hint, things have gotten a bit more advanced as well as a bit prettier)

 

It would be a very long time before 'another' upgrade will be done, careful what you want to be limited by

 

I guess some people have entrenched opinions on this , I guess we shall have to cordially disagree 

 

Cheers Dakpilot

[Brano]

I dont understand this DX witchcraft at all,but I would say it is about graphics mostly. As much as we all would appreciate more eyecandy stuff, the principal problem with DN engine remains in AI and game physics running in one thread. This must be separated,as Danil stated in one of his replies,if we want to see increase in game performance. In DCS,it was in fact graphics part of the game,EDGE engine,which has been updated. No idea how their physics and AI computing works over there. In older sims,AI was mostly scripted and followed predefined behavioral pattern. In nowdays CFS and other games, AI is making decisions of its own,following complicated algorithms. I watched a video recently from the game Im interested to purchase in near future, Kingdome come: Deliverance. One of the devs was explaining,how NPCs "think" what to do and how to behave in numerous situations. Those lines of code and flowcharts were mind-boggling 

[Dakpilot]

Although DX12 is graphics API there is more to this,  before your graphics card renders any scene, the CPU needs to simulate the in-game world, then prepare the instructions for the GPU to draw the scene. The more complex the scene, the more 'draw calls' are prepared by the CPU. 

 

 

DX12 allows CPU to perform 10 times more draw calls and fully utilise multi core CPU, DX9-DX11 is pretty much restricted to a single core to supply GPU with info, hence bottlenecks = stutters when this is maxed

 

Freeing up CPU for AI/FM and environment calculations is very important for Flight Sim, better performance on lower end machines = more players =  more theaters, features and aircraft

 

It is not witchcraft   there are some very important overall performance improvements, and when combined with current hardware there are many more performance benefits to DX12 tech as well

 

Cheers Dakpilot

[Brano]

When I fly alone,the game is smooth even on my obsolete CPU. It starts to get choppy when more AIs enter the scene. Thats why separate physics and AI calculations should be priority over DX12. It will bring increase in game performance for everyone,not only those having W10 and DX12 compatible GPU. DX12 is not a panacea for this game 

[6./ZG26_5tuka]

FMs/physics do not effect FPS and thus cause strutters as confirmed by Han. It's graphical feautures such as objects popping in within render range, animations as well as Ai and mission logic that causes biggest perfromence strain.

 

Graphicly BoS has been cut down quite a lot to compensate for that but obviously that's not solving the main issue.

[Dakpilot]

What are some of the most maligned limitations in BoS and many other sims?

 

Pop up of forest and other objects, aircraft visibility range, moving grass bubble, horizon "fog" low FPS at ground level, Cloud/smoke particle slowdowns and pop up, low vis of targets from bombers etc.

 

How many complaints were there when there was a bug in ground texture quality/visibility range, introduced while trying to improve performance/visual quality balance?

 

all of these immersion killers are related to DX9 limitations, DX11 does not address this issue sufficiently,  DX12 would allow a significant leap forward in these aspects whilst freeing up CPU load

 

IMHO ( ) an expensive engine update only to DX11 would not bring enough benefits for the future and be a huge opportunity missed, in one year's time when consoles are fully using DX12 performance

 

BoS in DX11 on PC will seem very old hat to new customers

 

Cheers Dakpilot

[Brano]

That's all true but without possibility to get larger amount of AI into air/on the ground,it is still only eye-candy stuff not adding to the CFS gameplay much. I'm not against DX12, but first things first

[6./ZG26_5tuka]

As far as DCS goes they sayed they've rewritten the engine code along Dx11 implementation to move rendering power from CPU to GPU and thus freeing the CPU up for different tasks. The result was an increase in graphical quality and performence (40 vs 120FPS in my case).

 

The graphical cutdowns of BoS are not because they are the limiting factors of performence but because they had to increase performence sacrifised for other things.

 

In the end this choice lies with the devs. They've more valuable data than Steam surveys or performence speculations at hand allowing them to have a better judgement and their personal interests.

[Jade_Monkey]

We will always have people who dont want to upgrade their hardware or OS but that should not stop progress. Otherwise we would be stuck in the middle ages.

[6./ZG26_5tuka]

If progress means having to deal with a buggy and (to some hard/software) incopatible OS with enforced updates and hard drive scanning I'm fine being against progress.

[Jade_Monkey]

Nothing buggy here. Runs buttery smooth.

 

Im willing to bet that most of the incompatible items are pretty old (although functional). Im not going to explain how technology works because you probably now better than i do.

[SharpeXB]

Looking at DCS Dx11 will surely be capeable enought to turn this sim engine into a smooth and stable platform.

Since these are completely different game engines it's hard to speculate.

And although EDGE seems to run fantastically and look great as far as I've experienced it (On a 4K monitor and I haven't tried the Nevada map yet)

VR is where EDGE finds its performance lagging because DX11 still relies on the instructions sent to the CPU and the huge number of objects DCS renders just crushes the performance. Remember that 3D requires every scene to be drawn twice. So DX12 is about mandatory if we want to see flight sims run really well in VR. You can read the reviews and explanations over there but DX11 simply can't provide a quality VR experience.

 

I'm a complete computer dummy and I managed to upgrade to Windows 10. Everything works fine.

I did only do the update because it was free, otherwise there were actually not that many direct benefits. But one of those benefits is in fact being able to run DX12 whenever that comes along.

Edited June 27, 2016 by SharpeXB

[II/JG17_HerrMurf]

It is quite odd to see the tinfoil hatters in a technology driven game forum.

[216th_Lucas_From_Hell]

More so for a game which has users constantly connected and monitors their in-game activity to gather actionable customer data. Though that did stir some drama when Rise of Flight was just out.

[Picchio]

That's all true but without possibility to get larger amount of AI into air/on the ground,it is still only eye-candy stuff not adding to the CFS gameplay much. I'm not against DX12, but first things first

 

That's all there is to say, really.

Golden words.

[SharpeXB]

That's all true but without possibility to get larger amount of AI into air/on the ground,it is still only eye-candy stuff not adding to the CFS gameplay much. I'm not against DX12, but first things first

Wouldn't DX12 actually help make that possible since it can reduce the load on the CPU?