The online racing simulator
iRacing
(13603 posts, closed, started )
Quote from PMD9409 :Doesn't always take money to produce a good physics model though. I think we all agree that it is because they are Americans.

I think they've started to believe their own publicity and that is fatal.

You can see that Dave really does believe what he is saying at 9.45 in the below link

http://youtu.be/5gFb1Nyg5ZA
Quote from menantoll :I think they've started to believe their own publicity and that is fatal.

You can see that Dave really does believe what he is saying at 9.45 in the below link

http://youtu.be/5gFb1Nyg5ZA

That's just a PR video. He's not going to act like it is crap.

However him taking 2 weeks off after the "2.0" release was funny IMO. They say some ridiculous things though, I'll give you that.
I think it is perfectly understandable to take a week or two off sometimes. You guys in america don't have a lot of vacation days anyways compared to the rest of the world

E: I'm actually quite worried with something he said in the vid. He said that stewart drove some laps and was faster than their own test drivers. How slow the iracing beta testers are when someone can come sit in a sim and do faster laps immediatelly? I hope that was just marketing talk there instead of something else.
Quote from PMD9409 :That's just a PR video. He's not going to act like it is crap.

However him taking 2 weeks off after the "2.0" release was funny IMO. They say some ridiculous things though, I'll give you that.

Dave's a developer though, not an actor. I really do think they have started to believe their own publicity. Just my thoughts, I am aware I could be wrong.

Quote from Hyperactive :E: I'm actually quite worried with something he said in the vid. He said that stewart drove some laps and was faster than their own test drivers. How slow the iracing beta testers are when someone can come sit in a sim and do faster laps immediatelly? I hope that was just marketing talk there instead of something else.

It does make you think doesn't it, when they say that real world drivers normally have a problem with "getting the feel" of sims straight away. If it was marketing talk who was it pointed at? The sim crowd are going to say "are your testers not up to it? If a real world driver with no sim experience beats your testers so easily" or the GT/NFS crowd "Wow a real world driver showed those sim testers how real it really is" Or is the iRacing PR department not really up to the job?
Quote from Hyperactive :I think it is perfectly understandable to take a week or two off sometimes. You guys in america don't have a lot of vacation days anyways compared to the rest of the world

All the more reason to stop taking vacations, he lives in US, no vacations allowed.
Quote :
E: I'm actually quite worried with something he said in the vid. He said that stewart drove some laps and was faster than their own test drivers. How slow the iracing beta testers are when someone can come sit in a sim and do faster laps immediatelly? I hope that was just marketing talk there instead of something else.

It's well-known that their testers actually suck major balls. Since the DWC guys are just a marketting tool, they don't allow them to test. I don't know who actually get to do the testing, but I do know that most, if not all, are average or below average in driving standards.

Quote from menantoll :Dave's a developer though, not an actor. I really do think they have started to believe their own publicity. Just my thoughts, I am aware I could be wrong.

Of course they have. Every statement they make is in PR format. Notice how the smaller guys (Eric and Shannon, or even Steve) are a bit quiet on the forums when it comes to some things, and Tony is the guy doing all the big "PR" posts.

All I gotta say is that I called this way in advance (early/mid 2009 I'd say), so I win.
Site down for Maintenance?
Ford GT update, massive digging just to be able to find this crap.

First started with this thread, where racers started to question the tires on the Ford GT.

Then, someone emailed Robertso ... s the real life Ford GT).

The email sent:
Quote :As you probably know, iRacing finally released the Ford GT to the masses yesterday. I aboslutely love the car. I was wondering though, does the team have the ability to change the steering ratio? The iRacing car has an almost F1-like steering ratio making it hard to handle for some and it can not be changed. Thank you, Patrick Lindsey

The reply:
Quote :Hi Patrick,

I haven't actually had time to even look at the iRacing version of the car yet. We allowed them to do a laser scan for iRacing so we could use the scan to create a CFD model for aero testing. Still working on that...

The actual car has power steering with electrical gain control. We cannot change the ratio while driving, but we can change the force required. In truth though, we set it once and never touch it. The car is probably more stable the simulated version.

Dave Robertson

So they did it so they can get the aero modeling from iR? Does that mean iR has to pay them or what? Anyways, to continue...

Staff member Ian Berwick, don't know him? Many don't, he's a quiet one. Well he just posted some things about the Ford GT that can be found here.

Here is what he said:
Quote :The steering ratio is not variable, so the 'quickness' remains the same regardless of track. The power assist is what they can increase or decrease.

We try to model the car so that it matches the original as best we can with the information we have available to us.

I did get additional (OK--the only ) aero data from the team when they were at Lime Rock in early July. I STARTED trying to get data before Christmas. Sometimes it happens quickly and painlessly, but this was more like pulling teeth. And I'm not done with the aero yet. I'm hoping to get some CFD done on this car 'soon,' whenever that may be!

And please don't bother the team with questions! I don't think they have the time or the inclination to answer setup questions. The spring rates are ranged according to what the engineer told me.

Shifting is now via pneumatic shifters. The old manual shifter is still in the cockpit, so they can get back to the pits should the pneumatic system fail, but that's what's being run here in 2011.

The tire model (we know) needs to run unrealistically high tire pressures right now. I believe we posted comments to that effect when the NTM was rolled out.

There is an entire laundry list of stuff we're working on(across the board), that take time with all the details and interactions that are involved here! Especially in the tire model!

He also mentions Dave's upcoming post:
Quote :I know Dave has been working on a post now that he's back in the office! He spent all day yesterday going through the forums, and making a multi-page list of things to say... we'll see how long or short the post is when he gets it finished up!

Funny how that is the first time that the Ford GT having unrealistic tire pressures on purpose, and the staff member said it thinking it had been told to the community before. Rather humerous.

lol his response to my post saying that:
Quote :And I'll probably get yelled at for doing it, too! LOL

Quote from PMD9409 :So they did it so they can get the aero modeling from iR? Does that mean iR has to pay them or what?

Iracing is not a company that can do cfd analysis. What they can provide is a millimeter accurate 3d-model of the actual car that the team can give to some company (along with other telemetry data they have for example) that does that sort of thing.

Iracing most definitely is not capable to produce any "aero modelling". Iracing relies just as much as some racing teams to getting that data from somewhere else.

It is just a deal where both have something the other may find useful. Iracing gets the car, robertson racing gets a laser scan of their car. Among other things I'd assume.
Finally a word on tyre pressures, it fits what we all assumed but why couldn’t they have admitted this in the release notes or the first few days?

Probably because we would speculate that other factors in the tyre model were severely fudged as well!

Well there strategy back fired as there has been intense analysis by the community and they have concluded that themselves anyway
Quote from Glenn67 :Finally a word on tyre pressures, it fits what we all assumed but why couldn’t they have admitted this in the release notes or the first few days?

Probably because we would speculate that other factors in the tyre model were severely fudged as well!

Well there strategy back fired as there has been intense analysis by the community and they have concluded that themselves anyway

Best part is when the staff member acted like they had already mentioned it to the community.

When I see him say "Don't contact the team anymore..." I look at it as "Don't contact the company, you might find we have many things fudged".
Heh, that is interesting but I bet Ian's purpose in saying that is more along the lines of "don't feck up our relationship with the team by pissing them off via flooding them with questions they don't care about"
I'll be sure to email the team now to see if they run a minimum of 35psi or not, because the new revolutionary sim has that.
I sense some frustration. Why bother?
Quote from cargame.nl :I sense some frustration. Why bother?

As bad as it may seem if it wasn't for iRacing I'd very likely have dismantled my racing rig by now as it would have been a waste of space for the last two years
Quote from cargame.nl :I sense some frustration. Why bother?

It is mainly frustrating because they have the resources to create something amazing, but don't strive to do so. I personally love racing, and after not being able to continue real life racing due to funding, sim racing was my only choice. It's frustrating to sink money into a sim with epic content, but the physics of a highly polished rfactor mod. Meh, I'm on my phone in class so i cant go on a proper rant, this should suffice for now.
Quote from PMD9409 :snip

What happened to you Phil? You used to be so good at your in class rants
Quote from Seb66 :What happened to you Phil? You used to be so good at your in class rants

Class was ending. Didn't want to type a long rant while walking across campus.
Release Notes for August 29th. Maintenance timing TBD

Quote from Greg Hill :


HPD ARX-01C

- Now has all basic and fixed setups for all tracks used by this season's series.



Ford GT

- Reduced braking power to allow more brake pedal travel.

- Increased power steering strength to help with the heavy steering.

- Increased front downforce by 10% to help limit nose light conditions, after re-analyzing wind tunnel data.

- Now has all basic and fixed setups for all tracks used by this season's series.



Chevrolet Impala

- Now has all basic setups for all tracks used by this season's series.

- During this season's update a change was made that made older replays not play back properly. This has been fixed.



Chevrolet Silverado

- During this season's update a change was made that made older replays not play back properly. This has been fixed.



Simulation Performance Improvements

There are two new memory related settings in the advanced graphics options, one related to GPU memory availability (video RAM) and the other related to CPU memory availability (system RAM). This change is intended to improve the simulation's scalability, and to help address some of the current memory related performance issues in multi-class sessions.

There is now a new performance meter which helps diagnose memory related issues. It is attached to the L/Q/S meters, and the new meter is named P for "page faults". More information on this new meter follows.



GPU Memory Setting

The "available texture memory slider" is now "Max GPU memory to use", and it now limits how much video memory the simulation will use total rather than how much it will use only for textures. If you have more than one GPU (SLI or Crossfire), set it to the smaller one's memory capacity as resources are duplicated on each GPU.



Max System Memory Working Set Setting

The new "max system memory working set" setting has been added to the graphics options, below the GPU memory slider. It defaults to 1200 MB for systems that have 4 GB or more of system RAM, and 1000 MB for systems that have 2 GB or less. This setting controls (roughly) how much system memory the simulation will use at most. If you have a lot of RAM (4 GB or more) and never have performance issues, you may want to increase this limit a little bit from the default. See the Q&A section below.



Page Fault Meter

A page fault indicator, "(P)age Faults", has been added to the (L)atency/(Q)uality/(S)kew meters. This new meter shows hard page faults and soft page faults as different colors.

Hard Page faults are by far the worst type, and show up as green, yellow, or red, depending on how many are occurring per second. If you see temporary green/yellow/red spikes on this meter it is normal for most PCs. But, if you see constant levels of yellow or red page faults, you may be suffering from a lack of physical memory. To remedy the situation either lower the "max system memory working set" slider in the advanced graphics options to reduce the amount of memory required for the event (and restart the sim), or else it may help to add system RAM to your PC. This meter currently tops out red, at 30 hard page faults per second. A steady stream of even a few faults per second may slow the simulation down significantly as your disk drive is being used as system RAM, so they are to be avoided if possible.

Soft page faults are indicated as a gray bar on the same "P" meter, and are not nearly as serious as hard page faults. However, they are displayed to you as a warning, as they may indicate that either the GPU memory slider, or the CPU working set slider is set too large. If you see a solid gray bar in the "P" meter, try to eliminate it via options, but if you can't, don't worry about it too much, it may be normal in some situations.



Q. What do I set the GPU memory slider too?

A. As a starting point, set it to your GPU's memory. If you have SLI or crossfire, choose the smaller GPU's memory as resources are duplicated on each GPU. If this setting is causing performance problems lower it slightly and retest.



Q. In testing sessions my textures look fairly high-res but when I enter an on-line multi-class session some of the textures look very blurry?

A. The two new memory settings cause this effect. When large tracks and/or more cars types are loaded in a session, more vertices are placed into both system memory and video memory, leaving less space for textures. The texture quality is automatically reduced to fit them into the remaining space. The limiting is necessary to avoid exceeding your PC's capacity.



Q. Cars and trackside objects look like I have them set to MED or LOW detail, but I checked, and they are still at HIGH detail in the options?

A. If the current working set (for the track and cars) is predicted to be way over your specified limit during loading, the simulator may now skip loading the top levels-of-detail for some objects to help reduce the working set to make more room for textures.



Q. I never have performance problems and after the update my textures look very blurry at some events. How do I make it how it used to be?

A. You may be able to adjust your settings to reduce these effects. During the event, go to the advanced graphics settings and look at the reported load of video memory and system memory working set (next to the two new sliders) to determine which setting may be causing the issue:

(A) Check the reported system working set load. If it is approaching the specified limit (within 200 MB or so) then it is likely this setting that is limiting the texture quality too much. You can try increasing the working set slider by 100 MB or so, and then restart the session. Do not raise this setting if it is not the limiting factor. If you start seeing a lot of page faults now on the new "P"age fault meter, or if your performance degrades, then you should probably reduce this setting. Raising this setting too high will result in performance loss, and often will result in page faults. We do not recommend setting this value above 1100 MB for systems with only 2 GB or memory or less. If you have 4 GB or more system memory you may be able to raise this limit substantially without performance problems, but do so carefully. You need to restart the session for the changes to take effect properly, and to test the new settings. If you set this value too high your textures may actually get worse and/or the sim may fail to load at all.

(B) Check the current GPU load is near the specified GPU memory limit (within 100 MB or so) then it may be the setting that is limiting the texture quality too much. In this case you might try to raise it a bit and restart the session. Don't raise this setting if it is not the limiting factor (as it will have no effect), and don't raise it too far. Setting it too high may cause the simulation to fail to load, page faults, worse performance, etc. Generally we do not recommend setting it higher than the actual GPU memory capacity. A different, safer, option to improve texture quality when limited by video memory is to disable some other users of the video memory. For instance, turning off graphics options such as "depth of field" and/or "shadow volumes" (in BOTH replay and graphics settings) will free up a substantial amount of video memory, allowing better texture quality. Selecting "cache swap nearest cars" in the graphics options may help as well when limited by video memory. You need to restart the session for the changes to take effect properly, and to allow you to test the new settings. If you set this value too high your textures may actually get worse and/or the sim may fail to load at all.



Q. I am now having new performance issues after the update?

A. This may be due to the new GPU memory setting. Try lowering your max GPU memory slider from the actual amount of GPU memory to a lesser amount to see if it makes the issues go away. If this doesn't help, put it back to the correct level and try lowering a different setting.



Q. I am getting solid yellow or red bar in the new "P" meter"?

A. Your system is using your disk drive to make up for lack of available physical RAM. The easiest solution is to lower the working set slider in the advanced graphics options to make the simulation use less RAM. If this doesn't work you may benefit by adding RAM to your PC. Other processes and services running on your PC also use RAM, so it can help to reboot your PC and close down as many application as possible before running the simulation - to free up as much physical memory as possible.



Q. I am getting intermittent flashes of green/yellow/red /gray bars on the new "P" meter, but it isn't constant?

A. This is normal, as long as they aren't associated with huge stutters.



Q. I am getting a solid gray bar in the new "P" meter?

A. Your system is incurring a lot of interrupts to manage virtual memory. You may benefit from reducing your GPU memory slider and restarting the sim. If this doesn't help try reducing your system working set memory slider. Soft page faults may not be that bad, but if you can get rid of them, it will surely help performance.



Q. I want to raise my working set beyond 1536 MB but it won't let me in the options screen?

A. You may override the limit by setting it directly in renderer.ini. This is not advised.



Q. I have 2048 MB of video memory but the load never exceeds 800 MB?

A. Due to the possibility of needing to restore a lost Direct3D device (i.e., if you alt-tab in full-screen mode), the simulator keeps copies of textures, vertex buffers, and index buffers in system memory. The video driver also uses system memory in its operation. If the simulator's predicted working set would exceed your specified maximum, the simulator is forced to reduce texture and geometry quality to keep the system memory working set in check. Raising your working set setting may allow your video memory load to increase (after a restart of the session), but doing so may also result in page faults and/or decreased performance for the CPU if set too high.

I really don't understand, first when iR opened to public EVERYONE who played it was like 'Best sim evahh!! oh yeahh' and now like 90% of this forum bitches about the game saying it's not the best and it sucks and shit...
I bought it because you guys said it was far better than any sim out there, even better than nkp, and it seems that right now it isn't?
Too much fudge for my liking.
Quote from chanoman315 :I really don't understand, first when iR opened to public EVERYONE who played it was like 'Best sim evahh!! oh yeahh' and now like 90% of this forum bitches about the game saying it's not the best and it sucks and shit...
I bought it because you guys said it was far better than any sim out there, even better than nkp, and it seems that right now it isn't?

I've had alot of hate toward them for over the past year now. Many have heard me go on long rants about them. Their promises in beta still haven't fell through. It just gets hold how they try to promise all these different things, and every single bit of it falls through. I don't mind a sim like LFS that already has a fantastic base and then takes the time to redo most of it's model to produce something even better. However iRacing hasn't even scratched the surface of a proper physics/tire model IMO. If it wasn't for the tracks, I'd call it something like rFactor. This 2.0 release marketing crap is just like a shot in the dark for them to try and redeem themselves. However they redeem themselves by coming out with something we would see from a console series.

If it wasn't for the content, and by that I mean the detail of it, then I don't see why many would stick there.
Well now you're getting carried away again Phil...

iRacing is still a better sim but has core faults that frustrate me given the resources they have, and have made use of. I'm still really waiting to see how their NTM works on road cars because that will tell a lot more than putting it on much higher performance cars - man what a mistake. As has been said, the Skippy is very good now and they should've worked their way up the scale instead of putting the NTM on demanding cars with known deficiencies that could cloud the issue.

Listen, I ran LFS last night for an hour because it's still fun - yet I was taken aback by this slowly oscillating "bouncing" effect that the cars have when subjected to sustained lateral load. What the hell is that? if iRacing did that I'd be even more pissed than I am at them for making great claims that didn't come to pass for the most part in their purported miracle build. That's pretty distracting issue now that I've been running iRacing which doesn't do that!

By and large iRacing still drives more like a car than LFS does, though often feels less natural - if that makes sense. I first thought that statement was paradoxical but it's not. iRacing can feel fudged where LFS can feel just plain goofy. LFS still has a better approach and a more consistent physical feel to it, but it's just too far off to be believable right now, even though that implicit coherency lends itself to simpler immersion. iRacing is rougher around the noticable edges but generally finds itself in a more believable input reaction - kind of like the "kid rails at a bowling alley - the ball goes where it should even if it's awkward".

Things are at a bit of a stalemate but even though I feel a little betrayed by Kaemmer's nomenclature and bravado, I still think they do have the ability to pull this off. I forget who said that they're in essence not using their dough properly (probably Phil) but the song isn't over yet. And if one company has the NEED to do this right it's them.

They really should buy Scawen out for some insane cash that sets him for life and use his talent. They won't out of pride (there's your Americanism Phil) but that really would be the best for the product and the market. Give the guy a few mil, move him to nice house all expenses paid and profit from one of the best minds ever to grace sim-racing - and I think he tops Dave considering what he's done on his own.
When you drive the fast cars to the absolute limit, or even just within 90%, you begin to see the amount of fudge that goes into that sim.

You are correct that I was the one with the spending debate, and also Americanism, so true.

But back to serious mode before bed. Lets just take the past 2 days into retrospect here. If you look back in this thread about 10-15 posts ago, I put quotes/links to where someone contacted Robertson Racing about the Ford GT. Now notice how they had not got the laser scanned car in order to get more aero data. Yet, iRacing is now getting more data to re-analyze and be able to "improve" the Ford GT. Of course this goes by adding a straight on the money 10%. Not 5.2, 6.8, 8.134235, no no no, lets go with a flat 10%.

Does this fix the problem with the rear of the car? Or does it just make it where the front of the car now can't get above the ground either?

Sure LFS might not have the best road cars, and that's where iRacing shines a bit. But even the other spreadsheet sim (rF) had a great Skoda mod, which was low powered road/race cars.

The live tire models seem to have a problem with low speed grip, which can be seen in LFS. Road cars in LFS have low grip which result in low speed, and you mix that together and you have some "meh". But IMO, TBO racing is some of the best racing out of any sim.

Looking at LFS' slicked tire cars (besides the MRT), they seem to handle much more proper compared to the ones in iRacing. A real life racer of a C6R was in the forum saying how a C6R compares to iRacing's IndyCar. It has that much grip, yet the one in iRacing is a drift toy.

iRacing just goes by that motto "You might be able to talk the talk, but you can't walk the walk".

BTW if you are oscillating in LFS, you are probably just moving your wheel. I get that in my car though when going over hills, is that what you mean?

Quote :iRacing is still a better sim but has core faults that frustrate me given the resources they have, and have made use of.

See, that is the statement right there that gets me irritated. How can it be the better sim if it has core faults? A sim is about physics of the cars so we can simulate real life racing. If the core has faults, than the physics have faults, therefore it is not a proper sim. "But it simulates real life racing like GrandAm, NASCAR, F1, etc". Well duh mother f****r so does NFS, GT5, and Forza, but that doesn't mean they are simulators.

I see it as selling a car that has engine problems. It sure shouldn't sell for retail price, or more than its competitors. But I guess the wheels, leather seats, facebook integration into the radio display, and many other useless functional features are more important.

lol I'm going to bed, that rant sucked. Sorry for those that read it.
Rant wasn't bad.

What I was talking about when I said that iRacing is a better sim despite core flaws is really due to the fact that it just seems that that is literally the case. The end result is more correct for the moment (settings and pressures notwithstanding I might add, that's more icing on the cake at this point for either sim) in iR than in LFS ONLY in terms of reaction.

LFS has core faults too, really really big ones some of them similar to iRacing ironically (at least with the NTM)

Like you I don't care about the series (unless everything else is perfect, then all of a sudden it becomes cool) I'm still talking about the basic driving experience, and both sims seem to be closing in on the prize but from opposite directions. My rants about iR lately are mainly due to the fact that my expectations were high given their obvious advantages as a company.
Quote from Ball Bearing Turbo :

By and large iRacing still drives more like a car than LFS does, though often feels less natural - if that makes sense.

It makes a lot of sense. iRacing cars behave very accurately. The Skippy is downright spooky to drive because it is crazily realistic in it's behavior.

But putting a cardboard box on some casters and rolling down a hill in it would feel more like a car than iRacing.

iRacing cars just do a terrible job of communicating what the car is doing. The audio is fine. But the force feedback and visual feedback (head bob) is terrible. The head "physics" only communicates bumps, but it needs to communicate G forces as well. The force feedback just goes slack when the car starts to really work, and doesn't warn of impending slides at all. It's like there is no self aligning torque to transmit and all the force in the wheel is coming from the caster.

That combined with dodgy slide behavior makes it too unforgiving and a chore to drive. I haven't raced for two weeks - I just don't feel engaged by the sim any more.

LFS, Net Kar Pro, and surprisingly GT5 all give very convincing feeling from the cars and actually feel like driving, even if there are problems with all 3.
This thread is closed

iRacing
(13603 posts, closed, started )
FGED GREDG RDFGDR GSFDG