Discussion in 'News & Notifications' started by Christopher Elliott, May 23, 2017.
Should be here:
yeah thanks, thats where I am, mustnt be up yet
Congrats to everyone @S397 on another flawless qualifying round today. Very impressive to see all the new tech needed to host this working so well in the background. It gives rFactor 2 great exposure for sure.
I was wondering what platform will be used for the finals of the MWFG. Is it already decided?
Very proud here of Avid Chronic Racing drivers Martin Stefanko Michi Hoyer and Teemu who qualified for Zaandvort.
Martin was P2, Michi P4 so we had 3 drivers in the race and 2 reaching the final so far. Teemu had good pace but quali didn't pan out and passing very hard here.
The cars were skinned nicely and ACR got mentioned in the stream.
Avid Chronic were the new boys on the block in FSR 3 years ago and have moved up the ranks organically by growing the talents of its members. It has an active driver development program.
To now have drivers achieving podiums in Pro, Ace and World Championship and have performances at this level in WFG has made us all very proud at ACR.
Familiar beginning, familiar finish... er, I'm talking about being back at Silverstone . What I notice as an aging old guy is that it takes longer for my muscle memory to kick in. Each time I come in to run a few laps, it takes a good 5-8 laps before I can be close to my previous session times and contemplate improving the lap times.
Median lap time is 1:02.05 for 317 entrants .
From 1:01 to 1:03, there are 187 people (59.0% of total).
To be in the top 10% of total entrants, you still need to be below 1:00.82.
50 people are in the 1:00s (15.8% of total).
Top 20 are 1:00.32 to 1:00.72, so expect 59s soon for these excellent drivers.
may I ask whether it was ever proven that this game behaves very much the same on all different hardwares possible (cpu, tact rate, memory, gpu, wheel hardware+driver ...) within the specification of minimum requirements?
Unless you are falling from real time it should behave the same. However, the higher the framerate and the lower the input/display lag will give you some advantage with respect to a struggling system. In order to be competitive you may have to reduce graphical settings (shadows, reflections and AA)
I asked if it is proven
@Stefan_L_01 the devs have said so, on multiple occasions. I tend to trust them.
It doesn't have to be proven. Fixed time step simulation algorithms already make sure that is the case.
A real time fixed time step simulation will provide exactly the same results (to the same inputs) in any hardware as long as the hardware is proven to give the same results to the same instructions and data (as in giving the same results when solving the same equations for example), and as long as those calculations take less time than the time delta between samples.
Intel and AMD already make sure and prove that their architectures produce the same results to the same inputs before releasing their processors, with the exception being bugs in the design which are very very rare.
So as SPASKIS said, unless your processor is so slow that you fall out of real time then yes, the simulation will produce the same results on different hardware.
I must say that there can be slight differences between AMD and Intel processors in simulation due to a difference in floating comma.
I happened to encounter this in my job when I was getting different results for the same simulation being run in different systems. Furthermore, in one case the AMD system would diverge and crash. It was quite a sensible simulation requiring really small time steps (1e-7s) in order to converge. Despite having a good time step strategy this type of simulation randomly crash sure to failing to comply with convergence criteria
However, for a simulation running at 400hz (2000 Hz for tire model) there shouldn't be any relevant difference.
At least according to my personal experience in transient analysis.
The only hardware hack affecting performance I could imagine is if the internal clock of the PC could be distorted. However I'm not an expert in real time simulation and it's just a wild guess.
Finally, it is possible to create cheats for rF2 by altering RAM locations where parameters are stored. At least it was possible until a couple of years ago. Actual system makes it very difficult for leagues to control this situations. Every build used to change this memory allocations so an anti cheat system would go obsolete quickly. The amount of work it was required to trace for all parameters that could be used for cheating was huge. For the cheater it was much easier since he only had to look for one or two variables to make the car quicker.
When that happens it is usually due to the use of libraries or compiler settings that execute slightly different code depending on the architecture where the code is running.
When running the same instructions they should produce the same result since they implement the same IEEE floating point standard. When they don't it is a bug and a pretty big deal like in the bug I linked in my original comment.
Median lap time is 1:02.00 for 353 entrants, little change from yesterday.
From 1:01 to 1:03, there are 204 people (57.8% of total).
To be in the top 10% of total entrants, you need to be below 1:00.80.
56 people are in the 1:00s (15.9% of total).
Top 20 are 1:00.30 to 1:00.60, field is tightening, but little progress at dropping fastest time.
So the answer to my question is simply: No
Where can I see my proc is "in time" ?
Afaik all gui output is gpu load
That would be wrong. If you want it in simple easy to understand terms, then the answer is still a simple "there is no need". Very different than a "No".
BTW, this is a thread about the WFG competition, if you need technical support post the question in the technical support forum.
Either the multitudes have realized they aren't the WFG or the circuit no longer appeals to them. Of those participating, there's certainly a bunch who have not improved their times which makes it hard to tell if they're no longer trying or if they've hit a wall like I have.
Median lap time is 1:01.99 for 388 entrants, little change from yesterday.
From 1:01 to 1:03, there are 228 people (58.7% of total).
To be in the top 10% of total entrants, you need to be below 1:00.76.
64 people are in the 1:00s (16.5% of total).
Top 20 are 1:00.26 to 1:00.52, field is ever tighter.
In my case this WFG is getting boring and tedious. There is no sense in having so many rounds. The interest for last round will probably be even less.
It's quite natural that less and less people will partecipate in the competition as rounds progresses .. it's a selection, if the last round only the 10 fastest guys will compete, it will be stil in the scope of the competition, so a success, because in the end, there will be only one winner anyway.
The actual number of people interested in and capable of winning this competition are only a small percentage of overall participants.
I just gave it a go for a few laps at Silverstone GP to see what it's like to actually press the button to get on that server but haven't bothered since.
Every time I tell myself I will go home and have another crack at it and when I do get home I end up doing other things.
Except for those competing for a McLaren job there's little of interest here to the rest of us.
I lost interest as i dont want to drive silverstone 3 times, especailly not the rubbish (imo) short layouts.
Separate names with a comma.