The netcode; what are your expectations?

All basic elements from AC we now know are even better than we expected. FFB, graphics, interface, even the sound is a big step up from netKar Pro.

The only big question mark that remains is the netcode. Traditionally the weakspot for Kunos Simulazioni.

I am curious what your expectations of the netcode of AC are. Do you think it will be comparable to iRacing's netcode for instance? How many cars do you expect to be able to race online?

What are key issues for the netcode to get right?

Shoot!
 
If it´s like Rfactor 1 where you could have half your car inside someone i might just pass.
Just look at Formula FSR, it just looks ridiculous how they get away with some overtakes since the cars just pass through each other.
It´s just not acceptable with the level of standard Kunos has set up for everything else.

If it´s anywhere near iRacing (which i think has by far and away the best netcode ever) then you got yourself a customer.
 
I am sure the netcode will be acceptable at the least and possibly even great. (like iracing)
Netkar Pro has decent netcode and I can't complain about Formula Trucks netcode at all...As long as they have found a way to prevent online mismatches everything should be fine and I'm pretty sure that it is important to Kunos as well.
 
Kunos track record with regards to netcode isn't exactly great and its the only thing about this game that worries me tbh.

It will be a do or die as the game is now too good to have mediocre netcode imo. Needs to be good.
 
it needs to be upto ISI/iRacing standards, simple as that! I dont mind if its a bit dodgy at the start and gradually made better via patches, but they have to get it right. They fail in this area and it could spell doom for them!
 
If it´s like Rfactor 1 where you could have half your car inside someone i might just pass.
Just look at Formula FSR, it just looks ridiculous how they get away with some overtakes since the cars just pass through each other.

rFactor netcode is amazing, that issue is due to the collision model, more so because modders aren't keen on spending hours to properly put the collision feelers in all the right places.

Open Wheelers are for the most part silly for that, tin-tops are much better ;)
 
Ages ago they were posting on twitter when they were trying some things out with the net-code/implementing it into AC I think Stefano said it would be better than NKP in regards to its net-code.

Thing is after all the patches NKP netcode was fairly decent allot of the issues were with older versions of the game and that it didn't work to well when you had people joining with +100 pings.

Rf1 and RF2 net-code can be good if everyone has a good ping but when people don't it gets a real mess especially with faster cars.

Then i racing always has the issue of people blinking on and off if there ping is high which to me totally ruins the game.

I think an easy solution for driving simulators to reduce issues is to have a more clear server option that admin can set , booting people whose ping's go above 130 or so or at least sending them to the pits when it does happen.
 
I don't think that kicking people with bad connections is a good thing. In racing sims, competitors are often a sparse resource and diminishing it is just killing the game. I think the way iRacing deals with this is not too bad since you can't do miracles when it comes to networking.

And as for miracles, a netcode that offers a perfect experience doesn't exist. Most people will always have 50-80ms ping to the server, meaning that between two racers, there will always be 100-150ms of lag when comparing what is happening on the screen of the first racer to what's on the screen of the second. That means that in a F1 race for instance, there can be a difference of up to 10 meters between perception and reality, that's why we will never possibly see the same overtakes that we see on TV.

You can't blame the studio for real world limits of physics and the structure of the Internet.
 
I think the way iRacing deals with this is not too bad since you can't do miracles when it comes to networking.

It often seems like The i racing way of doing it is terrible ( its more a "not doing it" ) it ruins races just having cars flash on and off at random , Obviously like you said user numbers are an issue. i racing is half dead as it is with the faster cars on the road side so I guess they don't have a choice but to let people race with awful pings.

In fairness to i racing allot of races go without hitch from net-code and for me its been more of an issue with the absurd damage modelling when you do tap. Also maybe i racing actually has really good prediction its just that when people get outside of the limits of that you get the flashing , which to me would be more of an argument for not letting people race with awful ping's and packet loss.

Most FPS servers or other timing based games that are competitive will set a upper ping limit for users. Street fighter even has a frame rate limit presumably due to the way it synchronizes clients.

In the end there are many ways developers can get around some aspects of latency its just a question of development time , cost and the end goals of the product.

There will always be some sort of compromise and mismatch between clients but I'm sure it can be vastly improved from where we are at now in most games.

"auto" booting people to the pit from server option suggestion is simply a quick and simple way to deal with the problem no where near the best.

Maybe over the next 20 years or so we will also start to see internet physical infrastructure improve father reducing latency especially between continents , I was reading a while back that the hardware is there it just needs to be implemented allowing 30ms ping from china to uk.
 

Latest News

Shifting method

  • I use whatever the car has in real life*

  • I always use paddleshift

  • I always use sequential

  • I always use H-shifter

  • Something else, please explain


Results are only viewable after voting.
Back
Top