Released Matsusaka Circuit v1.00 Now Available!!!

What is the difference between these?

 

10/10 for including ALL the layouts! Thanks.

I have reasonable performance--pretty close to the same as Toban, just for reference for those who are having issues.

I also get a bit of flickering on the white painted lines for the main start grid. Stands out because the rest of the textures are top notch. This track is both a testament to how good a track can look if attention is poured into it and a warning how much work is required to make a track look good. It's why the big studios have large teams of graphics people.

 

Thanks Woochoo & ISI

 

Oh yeah, what's with the strange track layout maps? Let me guess: the image of the Suzuka layout is copyrighted by the track owners. I guess Reiza is less concerned if that's the reason.

 

I noticed that too.

Looks like a great track but my PC had a couple of midlife crises as soon as I left the pits. I'll turn everything down to "GP2 VGA" level and try again.

 

This is strange, i just compared both and got a complete different result.

I have everything max/high, 25 AI, Camaro GT3, Practice

Matsusaka GP & INT: Framedrops in the pitlane down to around 20 FPS, very high drops between T2 and T8 with partially down to 2 FPS.

Toban Long: 60 FPS in the pitlane, on track between 80 FPS and 115 FPS.

//Edit: with track details to LOW the high framedrops between T2 and T3 are gone, but still there with track detail to MIDDLE.

By the way Reiza, isn`t the base of this track Kansai?

 

The gdb file needs to be tweaked. Just did a 20% length race on the West Circuit, it lasted 2 laps. Surely on the shorter circuit they would go a bit longer than that?

 

The reason why it's 10 laps as defualt is because the pitlane isn't really built for pit-stop races.
It might even be impossible to fit two wide shaped cars side-by-side in the lane.

I had widened the lane slightly already, to ensure there was at least a chance that two regular width cars could fit. But to make it any wider would be a diversion from what i perceived to be reality, which wasn't my goal for the track, generally.

10 laps may be too short for some, and I'm sure there are many cars that can run many more than 10 laps on a tank of fuel there, but I thought 10 would be nice enough for a short fun race. And 10 shouldn't cause any trouble with AI needing fuel in pretty-much any car, with such a tight pitlane where there would likely be congestion or worse. So, 10 was considered as a safe default value

 

Jeurgen, I went back to test. With EVERYTHING set identically, I get the same performance here as at Toban Long.

What I haven't yet tested (and probably won't bother) is whether Toban is as sensitive to car optimization. With a field of 20 ISI Nissan GTR's, the track is borderline performance, but very usable. If I load a less-well optimized car, performance plummets. If I load a poorly optimized car, it is unusable...stuttering and too-low FPS. I don't think Toban has that problem, but what if it does or does not? It won't change anything.

Of course everyone's hardware maxes out at different stress levels, but I personally refuse to change settings for one track, especially when the UI does not permit that. This track will only be able to be used by me with "light" graphics cars

The track is very nice, as is Toban, but let's be honest, neither is anything close to approaching the photo-realistic graphics available in other titles. If this quality of Suzuka/Matsusaka brings even the highest-end current machine to its knees (or only the best optimized cars can be used), the entire graphics engine needs a massive performance boost (unlikely to happen...we've been waiting several years), or, failing that, don't bother adding so much detail to the tracks themselves. I'd rather have simpler/closer to AMS-grade graphics, but more tracks to race on than fewer.

 

This is really strange...

This is a point i dont get anyway...when i fire up RRRE or pCars, don`t talk about the physics...but i never had problems with this two sim`s, to play them with full gfx settings, nor had to fiddle around with nvidia settings etc...

 

Those framedrops I would guess are caused by the AI cars, especially the pitlane one. Not sure how the engine renders AI cars that are boxed in garage. I think they should default to lower LOD levels, but not sure if it's a track setting.

 

Did it again, with the Honda NSX, as advised from Marc:

24 AI, everything max, execept reflections (low).

Before T2

View attachment 20196

After T2

View attachment 20197

 

So something is combining together to overload the system. Matsusaka with a light car, OK. Heavy car on a simpler track, OK. Heavy car at Matsusaka and even a top system is brought to its knees. If woochoo has already looked at this and optimized to the extent possible, then we know why Reiza choose a different graphics engine to go with GMotor for their next sim. The graphics quality here (cars or tracks) just doesn't justify the struggling performance.

And before anyone says it's the super-advanced physics that causes all the problems in rF2, they are a constant in the scenarios above that swing from 4 FPS to over 100 FPS.

 

The track detail setting are there so you can customise the track to your performance capacity. You are given the control to try to make it work for you.

I put more detail in it because I know people can use lower settings if their systems can't handle all of the details. On top of that, if certain details weren't in there, I know people would complain that they weren't in there. In theory, this should work for both groups of people.

Perhaps it would be nice if ISI could add some user-defined track-specific detail adjustments, similar to how they have added track-specific AI settings, but perhaps you could check to see if the lower detail settings actually make a difference for you first. Turning down to High will make a small difference, Medium will make a slightly bigger difference, and Low should make a bigger difference than both Medium and High.

 

Might just be the highest quality track for rF2 atm. Very well done, thank you!

I don't have a monster system by any stretch of the imagination

i5-4690K @4.5GHz
R9 290 @1100/1425MHz
8GB of RAM
5760x1080

...and I can still run this track with 24 (14 visible) cars with my fps >60 at all times, most of the time I get 84fps (locked)
Shadows and reflections are set to low while the rest is set to max/high.

Sure, on most other tracks I'm getting 84 all the time so it's a bit more detailed but as the author said you always have the option to lower the track details.

 

I am not willing to lower settings, sorry, and especially not when it would affect all tracks, which zero have a problem other than this one. And because rF2 doesn't look that great even maxed-out. Graphics are secondary to physics and other features for me, but still. If I want less sophisticated graphics I can play AMS with a wonderfully polished interface and consistently polished cars and tracks.

I love your work, woochoo, but as I said, if there is no potential to further optimize this track in a way that will make a significant difference, then its the cars and the core engine that need optimization. That's for me. Some cars work fine. Other people will have systems that can't even run the track at medium. This is all a normal part of gaming and evolution of quality and detail of environments. However, when other titles look even better and have higher performance, it reaches an inevitable point of comparison.

 

If you know you're about to drive on Matsusaka, you can take a moment to change the Circuit Detail one, two, or three notches lower. Perhaps you'll forget to change it back when you load another track. But then you'll remember and take a moment to reset the settings. The graphics won't become 'less sophisticated'. There will just be less objects.

I don't know if you have asked ISI for track-specific detail overrides, and I don't know if it's possible, but it might be worth investigating

Anyway, you have your options, and now I think I know what you want to do, or not do as the case may be

 

This isn't really a direct reply to you, but it's just jangled that nerve I guess. And it is probably all OT.

I have seen people everywhere on gaming forums use what seems to be the buzzword "optimise". It's as if every issue can be solved by implementing it, and than any issue with performance can be directly put down to lack of optimisation.
Game A runs different compared to Game B? "Oh, the devs just haven't optimised it yet". Ignoring the fact that the games are engineered on different code, engines, platforms, with different goals and directions about doing things, etc etc.

It seems like people would prefer to put poor performance down to lack of optimisation rather than hardware not upto the task they are trying to give it. I think I can even recall on this ISI forum someone with an old Nvidia 6800GT, hoping the devs would pull their finger out and put in the optimisations in rF2 to make it run better for him.

And again, I think I saw it in reference to the AI systems in both AC and rF2 recently. AC using player physics for all AI giving an even experience, and rF2 using older rF1 styled physics as using player physics for AI cars would cripple any system around. That immidiatly brought up the "see, shows how poorly optimised the rF2 physics are!" arguement, which just doesn't make any sense. You can't just endlessly keep optimising something till it runs exactly the same...unless the two bits of software were exactly the same...

You certainly can optimise things in different ways. If you find a way of simulating something that gives very close to the same outputs but uses simpler methods, it might be wise to consider using said methods. But if you optimise something by making a whole system simpler, which results in taking you away from the goal you were wanting to reach, that doesn't make sense at all in simulations. It would be fair to assume that any professional software engineer would know this...

As an example, Arma 3. "If we remove the bullet physics and interactions, we might save a few fps during an intense fight". Doesn't make sense in something like Arma though, trying to simulate combat.
For CoD though? Yep that could be argued to make good sense, battles aren't typically on massive maps, and better performance there would be appreciated by players more than "the bullet should have drifted by a few cm there due to the wind, etc".


Similarly, I read the same from people who worked on the Gamebryo based engine that Bethesda use in their games. Lot's of people crap on it for performance, glitches and graphics, etc, but it was one of the Obsidian developers who said something to the effect of "No other engine around would work as well to fulfill their goal. A large, monitored, complex open world full of stuff and random interactions with the ability to be heavily modded by players as well as extended upon by developers". He mentioned how they were able to interface some completely different dialogue system plugged onto the games original system and it functioned perfectly with minimal work.


So in the tracks case here, maybe there was just nothing more that could be done to optimise the track, save losing details around the track. Maybe he didn't want to lower his track poly count because it then didn't function as well, or the realroad didn't work as well, who knows. Tuttle mentioned it with Mores, because it is so small, they were able to spend heaps of resources making that track look great, getting each little bit perfect.

Which then brings me back here..."when other titles look even better and have higher performance, it reaches an inevitable point of comparison.". Comparing two different titles just doesn't work like that unless you ignore or don't understand what is happening underneath the title. So someone should leave their constructive feedback and leave it at that.
In this case, you have performance controls at your disposal that aren't hard to change. Your choice not to use them of course, but it's about now when I start getting the vibes like that old Nvidia 6800GT guy...

 

@MiniBull, game performance is not being only based in user performance controls but the game programming and GPU drivers aswell. NVidia profile (rFactor2 Mod Mode.exe, rFactor2.exe) is created 2011-09-22 and it is to that old Nvidia 6800GT guy Nvidia has released hole bunch of new drivers after release of rF2 and none of them gives you satisfying fps, scaling to achieve normal game experience with Nvidia cards. ISI has been asked for a long time to submit latest build of the rF2 executables to Nvidia so that an updated working game profile can be created but unfortunately ISI seem to have communication problem with Nvidia.

I have same performance as latest 3PA and ISI tracks with this track, just for reference for those who are having issues.

 

It's not all about game engine. A race track is mostly made of "simple" shaders and materials, but due to the size of the area it can get tricky. For example Matsusaka needs above 2 GB GPU VRAM to run smooth on Full details and no game engine in the world can reduce VRAM space requirement, as that is determined by the texture and object resolution. ISI/3PA tracks often seem to use a denser mesh and higher detail trackside objects than what other sims like AMS and AC uses. This is the limiting factor.