Released Zandvoort 2017 is now available!

That's funny cuz my 1060 doesn't bottleneck -at 3.0x8,I see zero difference between 8-16 on Any game I play.
Something else is going on in your system,is you mobo asus by any chance,they labelled some of their mobo as 3.0x8 or 16 but they actually run a 4x

 

The problem is not my PC, I wish that it was, but I have checked that it runs x16 2.0. Anyone can repeat the test of @DrR1pper and change their PCI-E from x16 to x8 and see the difference for himself in rF2 (with 20 AI cars). I cannot as I have PCI-E 2.0, so I can't compare with the faster speed of 3.0, but bus usage tells I'm getting bottlenecked. Also, other games are irrelevant, as they don't use as much bus as rF2, not sure how I can make it more clear. rF2 has PCI-E bus usage roughly five to ten times higher compared to Assetto Corsa or Automobilista.

Basically it's a waste of money telling people to buy the fastest CPU on the market for rF2 when the real issue is that many people are running 2nd gen i5/i7 chipsets with PCI-E 2.0. At moments when I'm bus limited (like at Zandvoort race start), I cannot gain any FPS by overclocking my GPU or CPU, which conclusively tells that PCI-E is the problem.

 

So you can only run 2.0x16 which is the same bandwidth as 3.0x8.
I have a 1060 i5 on z97 chipset,pls explain why I don't see any performance decrease or increase if I run my system at the same bandwidth as you (3.0x8) than I do at 3.0x16.
If rf2 was surely pcie limited,shouldn't I be seeing the same.
I'm trying to get my head round this because 2.0x16 has the same bandwidth as 3.0x8

What's your mobo cpu and gpu,what ram do you have ??

 

You are right. There is huge difference PCI-e@3.0 x16 vs PCI-e@3.0 x8. ( PCI-e@2.0 x16) in rFactor2. I run test later this evening with MClaren at Zandvoort and post pictures. I find it also irresponsible when some are telling to people to buy new CPU when that is not the case.

 

List your system specs along with chipset to see if anything is the same as stonecs,because again if that was the case,I should be seeing this limitation too,no?

 

That's a completely agreeable point,

 

I guess, for a beta version 0.2, a perfect optimization cant be expected. As Richard Hessels (the creator of that track) stated out, a lot of objects are still placeholder...

 

Asus X99 WS-E/10G,
CPU: i7-5960X@4.4Ghz,
RAM: 64 GB Ripjaws 4@3000 CL15,
GPU: MSI GeForce GTX 1080 Ti SEA HAWK EK X

MClaren at Zandvoort:

PCI-e@3.0 x16 > 155-210 fps


PCI-e@3.0 x8 (PCI-e@2.0 x16) > 115-165 fps


There is nothing wrong with my or @stonec`s chipset. I don`t know if gMotor is poorly optimized or intentionally added to work like it does.
To be 100% honest with you few years back I couldn't either believe that gMotor is working like it does and was very assertive claiming that PCI-e@3.0x16 vs PCI-e@3.0x8 has same performance but I was proven wrong.

 

Are you running the latest bios version for that x99 board,a lot of people were running 2.0 even though it says 3.0,until they updated th bios.
I'm not doubting what either of you say,I'm just confused why I don't have this problem.

 

@patchedupdemon Probably because you are not running the same graphic settings or the same number of AI, that's the only thing I can think of, as depending on how much you load the bus the bottleneck will be more or less. I tested with 20 AI (McLaren) at race start, as that gives a good idea of the lowest possible FPS, which is important.

Here's another DX11 thread where a third guy confirms our findings (bottom-most graph), showing 30% less FPS with PCI-E 2.0.

 

ASUS X99-E-10G WS has PCI-e@3.0 x16/x16/x16/x16 or x16/x8/x8/x8/x8/x8/x8. Yes I have updated BIOS.
Do you have sound card or something else connected to your one of PCI-e slots?

 

If this was a question for me,then no I have graphics card and wifi card,with an hhd.
Do you have a anything other than gpu in the main pcie slots

 

I'm going to hold out for a month & look at reviews of coffee lake
May get that or a 7700k

Was trying to resist but I can't help it these new things are shiny you know

Incredibly i7 2600k still decently sell on eBay so I'll sell my mobo/mem/cpu after upgrading

 

I know you are a member on iracing,can you run all tracks there with no worries,if so,then maybe I was wrong to point to your cpu,as really they are still good considering the incremental steps each new gen has made since yours was realeased.

 

Yes no probs on iracing , only some tracks at night ( fine in day) with full field on start ( if nearer the rear of grid) there may be few minor glitches

I think though it's because iracing don't release beta tracks things seem very optimised & high quality out the gate , but then obviously users have to pay for them if they want the track etc
Think I'm a tad spoilt by the quality of their tracks -which is the best thing about iracing imo

 

No

 

Actually they do, sort of



Maybe rF2 Zandvoort should have been released in a similar state at first, then the performance could have been higher.

 

Yes they call that a tech track
Basically unfinished , driving around in a laser point cloud but with track & track side barriers filled in

Long beach has a phoniminal track surface lots of character , maybe because it's a street circuit

 

That is arguably the best road track on iracing,yet they have said there's too much work involved to complete it,the amount they would have to charge to make a profit on it,they know people wouldn't buy itand the sheer amount of trackside objects would grind everyone's system to a halt,hence they will never release it in a finished state.
This is one case where there ethos has let them down,they could release it with the same standard track detail,but have a simplified trackside state,but they won't.
They like people to believe only the best will do,unfortunately that doesn't apply to the tyre,dynamic track or weather models

 

No idea why they don't finish it but the stuff outside the trackside barriers are trival to me
I don't really look at buildings etc whilst driving perhaps be nice for replays though

It's still an ultra high quality track to drive lots of little cambers and changes in track -very noticeable when trying to get a fast lap
In spite of people's disbelief in laser scanned tracks ( & the point cloud being utilised correctly ) they can not be beaten

Agreed their weather stuff is pretty much non existent