Atomic Systems

So I am guessing that a new PC is required; since all the new machines run Windows 10 does TS have any running issues with W10?

firetrap1 wrote:I see that they require a GTX 1080 video card for 'Optimal' settings. That may sting the wallet a bit.

Everyone is free to have their wallet stung or not.

To me, it means that TSW will benefit from such an uber card while TS2017 not so much.

firetrap1 wrote:I wonder what kind of hardware will be required to get the best out of 3rd party DLC in the future, seeing as route builders and scenario creators already like to push the current sim to it's limit.

Better one. Future hardware will not perform worse than present one (or else no one would buy it). At least some future DLC will certainly utilise future hardware.

david1 wrote:

firetrap1 wrote:I see that they require a GTX 1080 video card for 'Optimal' settings.

So a super computer required.

You are right if and only if 1 equals 3.

There are three levels of hardware described. Level 1 is 'required'. Level 3 is 'optimal'.

deltic009 wrote:The current game is a heavily reworked but poorly optimised piece of very old software running a specific game engine rather than an established widespread one. UE4 edition will be fresh out of the box, and like when TS was Railworks will likely be far more reliable and stable because it shall be as developed rather than being pushed up to and beyond its logical limits.

But we don't know what is round the corner, in a couple of years time we could get 128bit systems, PC being am to use 128GB RAM, then UE4 could be pushed beyond its limits, like the TSW video part 1 says, other games don't need to use miles of assets to call on, so this is all new ground, UE4 gets developed but at what cost in performance on modest PC, as the game developes and UE4 gets more complex then the ability of modest machines will decrese to a point where the user is either forced to upgrade or not increase there dlc.

david1 wrote:

deltic009 wrote:The current game is a heavily reworked but poorly optimised piece of very old software running a specific game engine rather than an established widespread one. UE4 edition will be fresh out of the box, and like when TS was Railworks will likely be far more reliable and stable because it shall be as developed rather than being pushed up to and beyond its logical limits.

But we don't know what is round the corner, in a couple of years time we could get 128bit systems, PC being am to use 128GB RAM, then UE4 could be pushed beyond its limits, like the TSW video part 1 says, other games don't need to use miles of assets to call on, so this is all new ground, UE4 gets developed but at what cost in performance on modest PC, as the game developes and UE4 gets more complex then the ability of modest machines will decrese to a point where the user is either forced to upgrade or not increase there dlc.

You don't know that it will happen just as I don't know it won't, but we are only discussing this because of the pretty unique free upgrade system that has been in place for 9 years - any other game and each would have been seen as a different piece of software altogether without backwards compatible DLC and expectations that your previous versions system would automatically run the next or the next by default.

In that way we've been spoilt but it certainly shouldn't be a stick to beat Dovetail with in my opinion.

You can't push anything beyond it's limit------the word means it's as far as it will go.

One thing worth mentioning is this: How much you spend on a graphics card depends on how much you spent on the display!

This does not depend at all on the game. A 2560 x 1440 display has about 80% more pixels than a 1920 x 1080 display. Than means that for the same quality settings, you need twice as much memory and twice as many shaders/channels/processors.

Since the bigger displays are also more expensive, the solution is extremely simple. If you find yourself forced to get a graphics card, go where the money for the display came from.

A salient reminder, I'm stuck with 1680x1050 still, usually run at 1080P for screenshots and videos of stuff online.

david1 wrote:

deltic009 wrote:The current game is a heavily reworked but poorly optimised piece of very old software running a specific game engine rather than an established widespread one. UE4 edition will be fresh out of the box, and like when TS was Railworks will likely be far more reliable and stable because it shall be as developed rather than being pushed up to and beyond its logical limits.

But we don't know what is round the corner, in a couple of years time we could get 128bit systems, PC being am to use 128GB RAM, then UE4 could be pushed beyond its limits, like the TSW video part 1 says, other games don't need to use miles of assets to call on, so this is all new ground, UE4 gets developed but at what cost in performance on modest PC, as the game developes and UE4 gets more complex then the ability of modest machines will decrese to a point where the user is either forced to upgrade or not increase there dlc.

128bit won't be happening anytime soon, not on a commericial average user level, probably another 15-20 years for it to be even realised.

64bit OS and hardware has really only started to be utilised and used on a day to day basis in the past 2 years, prior to this most people only had a 64bit OS which allowed more ram, but actual programs being coded and optimised for 64bit systems has only really started to dripfeed in in the last year or so. There isn't a huge catalog of game titles which use 64bit and utilise it properly. Keyword "properly" because most which convert to or use 64bit need to be recoded from the ground up, especially if they were originally 32bit or make a really terrible 64bit conversion and crash most of the time since they bolted on the 64bit code onto the 32bit code, which quite frankly doesn't work.

Even the newest consoles have only really just started using 64bit and actualy compatible coding which allows for more than one closed enviroment, e.g. the xbox one uses windows kernels and 64bit artitecture, same with the ps4 but it doesn't use windows kernels for obvious reasons.

The UE4 engine is capable of alot and it also if its optimised and coded properly it can run extremely well and produce stunning graphical fildelity as well as features, its just entirely down to weather or not DTG can do that at the end of the day. Posted system requirements are ultimately a guideline and nothing more and with proper tweaking and configuration of your system and proper maintaince you can usually get more out of it. There will always be hard barriers, like for example the requirement of a higher CPU speed or a graphics card RAM requirement, but at the end of the day no one forces a gun to your head and tells you to cash in on the expensive hardware to run newer software.

I've advocated and argued for many years when this discussion comes up that anything built in the past 4 years will handle a new engine better than the current engine and the UE4 engine even on medium settings will be still far and above what maximum settings are currently on the simulator. You have to remember, TS2017 is based upon a codebase which was designed for directX9, the limitations and lack of graphical optimisation from that era are still present and its why runs like glue on larger routes because the optimisation for textures and data for the game to run is based on old methods and not newer stuff and better compression methods. Most people are on DX11 or 12 if your in windows 10, there has been a mountain of improvement and changes since then which includes better Anti aliasing in form of TXAA , SSAO, shader 4 and 5, shadow and light rendering technology aswell as a list which would go on forever.

We all can't live in the Win XP era, its time to move along with the flow of technology and stop expecting developers to cater to systems which are wholly outdated by this point, its tough love, sure, but if a game or simulator stagnants to appease a certain market share which refuses to upgrade then it will never improve or be any better in the long run since it utlimately has to handicap itself and be limited to ensure it doesn't push the limits of the older stuff.

DTG should have dropped the upgrade once a year policy and developed a new engine years ago, they didn't and that is why its took so long to get a new modern simulator which utilises hardware properly, other companies have done in, like Euro truck sim 2 and its american counterpart aswell as farming sim and some of the newer flight sim games, they all have adapted and moved to new hardware and can utilise it quite well, there really isn't any excuse in 2016 to be still using coding which is more than decade old at this point tbh.

Standing ovation for crumplezone......hits the nail rather squarely.

I don't quite understand the point!

I'm sure we all like a game that looks more realistic and performs better! As graphics improve as does the demand for power of the CPU/GPU (etc).

Fincra5 wrote:I'm sure we all like a game that looks more realistic and performs better!

Certainly, but some are taking that to mean that any improvement will necessitate a hardware upgrade---not always the case. Improvements in the efficiency of any routine in any piece of software often mean that existing hardware doesn't have to work so hard to produce the required results on screen.

crumplezone wrote:DTG should have dropped the upgrade once a year policy and developed a new engine years ago, they didn't and that is why its took so long to get a new modern simulator which utilises hardware properly

There's that, and also the fact that developing a new game costs money, which means a solid financial base needs building before such work can be done--it may be that it's taken 10 years to get to such a state. Any financial backers also need to see a company that's organised in a sound way, rather than one purely *chasing a dream*

gptech wrote:

Fincra5 wrote:I'm sure we all like a game that looks more realistic and performs better!

Certainly, but some are taking that to mean that any improvement will necessitate a hardware upgrade---not always the case. Improvements in the efficiency of any routine in any piece of software often mean that existing hardware doesn't have to work so hard to produce the required results on screen.

crumplezone wrote:DTG should have dropped the upgrade once a year policy and developed a new engine years ago, they didn't and that is why its took so long to get a new modern simulator which utilises hardware properly

There's that, and also the fact that developing a new game costs money, which means a solid financial base needs building before such work can be done--it may be that it's taken 10 years to get to such a state. Any financial backers also need to see a company that's organised in a sound way, rather than one purely *chasing a dream*

Maybe, but from a games developement studio point of view, kickstarter and steam early access have been available for atleast 4 years as a point of drumming up support and funds for the final product, if done right early access can bring in alot of support and if you properly have a bussiness plan and show your doing stuff kickstarter can also cover the issue of funding.

A example of a pretty successful UE4 engine based game which went the early access route is Ark Survival Evolved, they were a pretty small studio which outsourced a bunch of work to other places and didn't start off with alot either. They had a working product, abeit abit rough on the computer systems at first, but they have been able to optimise that throughout the early access then release schedule. They now have console versions on the new consoles + originally on PC and doing pretty well for themselves.

Thing is, and this is my opinion solely, DTG didn't do the whole scrap and redo the whole simulator because they were in a safe monopoly position for the train simulator market, lets face it, there competition is MSTS or Trainz and both are pretty outdated by this point and the latter is very model railway-ish in the way it "plays" and didn't deliver anything spectacular with its "New Era" version, infact you find the same cabs and models used from all previous versions still in the new version of Trainz.

Quite honest, I think DTG shot themselves in the foot by prolonging Kuju's rail simulator as long as they have and there is a massive addon catalog which they are going to have to catch up to on TSW and they also somewhat limited to the potential market with opting for a US themed simulator release, which for me basically means a no buy for me to there is evidence of atleast some UK content. Its not I don't like US content, I've got plenty of it in the folder, but I'm not really interested in it from a driving experience point of view. Its kinda the same with me and euro truck sim 2 and american truck sim, I like both of them but I play ETS2 more as its well, more familar.

TSW, in theory, should run alot better in all points, as it should be using most of the new graphics tech and file structuring of modern games, which means better file compression and loading in real terms and in terms of graphics cards any of nvidias 7XX series or the lower 9XX series will be able to run it well and to a decent graphic quality to. Thing is aswell, graphics card wise, with the new 1xxx series from nvidia it means the prices on the 9xx series have dropped a fair amount, 950s I saw before black friday were sub £135 for example, in some cases below £100, the 970GTXs were also well below what I had bought mine which was £256 about 3 years ago. There is also budget 1xxx in the 1050s and 1060s, most people will actually be fine with the 1060, 1070s for more extensive requirements or if bat loco a 1080 which quite frankly a really crazy performance wise and are more suited to video rendering.

crumplezone wrote:with the new 1xxx series from nvidia it means the prices on the 9xx series have dropped a fair amount,

Definately---making claims that you need to perform a £700 upgrade completely baseless. 300 quid tops, unless you're really in the mood to help the country's economy.

crumplezone wrote:Maybe, but from a games developement studio point of view, kickstarter and steam early access have been available for atleast 4 years as a point of drumming up support and funds for the final product, if done right early access can bring in alot of support and if you properly have a bussiness plan and show your doing stuff kickstarter can also cover the issue of funding.

A example of a pretty successful UE4 engine based game which went the early access route is Ark Survival Evolved, they were a pretty small studio which outsourced a bunch of work to other places and didn't start off with alot either. They had a working product, abeit abit rough on the computer systems at first, but they have been able to optimise that throughout the early access then release schedule. They now have console versions on the new consoles + originally on PC and doing pretty well for themselves.

Kickstarter and early access also have a lot of pitfalls - not least having a rabid fan base (which I think train sim'ers very much are!) who can destroy the reputation of a product before it's even been released. It's also very hard to keep a steady hand on the direction of a product when thousands of voices are all screaming for their particular angle - and that happens a lot with train sim'ers as well due to our diverse tastes in locations, era and traction types. I get the feeling that train sims are too diverse a product to do well in the public arena.
And DTG have such a terrible public perception - some it deserved, most of it not - that I think throwing open a new project to public funding would have been a huge recipe for grief and anguish all round.

Also we should be grateful they waited until they had Matt P on-board to head it all. I can't think of anyone else I would you rather have in charge the next gen of train sim

karma99 wrote:Also we should be grateful they waited until they had Matt P on-board to head it all. I can't think of anyone else I would you rather have in charge the next gen of train sim

That's the one point that has not been questioned (yet??).

I agree on the points regarding crowd funding, too.

What is much more important: We don't have a say there. It is a company owned by persons who try to optimise their revenues, like any owner (or shareholder) will do, worldwide.

Since we don't have their business figures, we cannot even theorise about the best time to release the new system. I am sure they were working in the background in 2014 already, with some planning and testing in 2013. But who would kill a product that sells very well? After long years of struggle, the current sim got to a point where lots of people happily bought lots of add-ons with minimal complaint. As soon as the new thing will shine in full (which I expect in mid-2017 or so), everyone will feel bad about having bought "the wrong" add-ons which are then "totally useless".

People should thank DTG for releasing TSW only now and not earlier. If they release TSW earlier, the moaning about TS201x becoming redundant would have happened earlier, too.