If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#11
|
|||
|
|||
New PC's FFB
Mario Petrinovic:
Well, I got new PC, and there are some changes. On this PC I can clearly feel where I am regarding latency (cockpitLookDeadZone setting). Here, my latency definitelly is somewhere around 0.038500. Well, this has more sense than what I wrote previously. Thankfully now I can clearly distinguish where I am with my latency, so I can experiment with different settings (if I find time to do that). Regarding the feel of FFB, if latency is too low, your are chasing car, if it is too high, car is chasing you. When it is too low, car runs away from you, but you cannot catch it. This isn't so easy to distinguish, but it is very easy to distinguish too high latency. In short, with too high latency FFB behaves somehow like a yo-yo. Like it has a spring in itself, and this spring acts exactly like a yo-yo. It is controlable, but of course, because of latency it plays around, and in adjsuting to this, all feels just like a yo-yo. Well, now that I am sure what is what, now I can freely experiment with different video settings, to see how they affect latency. ---------------------- There is some development, here. Up till now I chased the divide between, yo-yo zone and no yo-yo zone. Well, this was wrong. On my system that divide is at cockpitLookDeadZone = 0.038570/0.038569. Well, the prefered value of cockpitLookDeadZone is deeply into yo-yo zone. I still didn't determine where exactly (I had enough of testing for today), but it is somewhere around 0.08. Good news is that it looks like it doesn't have to be so precise, it looks like three decimal places should do. Also, on my system the fps lock at 84 is too low, regarding FFB. 169 is good. 169 is better than "no lock". "No lock" is much better than 84, and almost as good as 169. I still didn't have time to see how low can I go regarding this, but 84 is clearly too low. |
Ads |
#12
|
|||
|
|||
New PC's FFB
Mario Petrinovic:
Mario Petrinovic: Well, I got new PC, and there are some changes. On this PC I can clearly feel where I am regarding latency (cockpitLookDeadZone setting). Here, my latency definitelly is somewhere around 0.038500. Well, this has more sense than what I wrote previously. Thankfully now I can clearly distinguish where I am with my latency, so I can experiment with different settings (if I find time to do that). Regarding the feel of FFB, if latency is too low, your are chasing car, if it is too high, car is chasing you. When it is too low, car runs away from you, but you cannot catch it. This isn't so easy to distinguish, but it is very easy to distinguish too high latency. In short, with too high latency FFB behaves somehow like a yo-yo. Like it has a spring in itself, and this spring acts exactly like a yo-yo. It is controlable, but of course, because of latency it plays around, and in adjsuting to this, all feels just like a yo-yo. Well, now that I am sure what is what, now I can freely experiment with different video settings, to see how they affect latency. ---------------------- There is some development, here. Up till now I chased the divide between, yo-yo zone and no yo-yo zone. Well, this was wrong. On my system that divide is at cockpitLookDeadZone = 0.038570/0.038569. Well, the prefered value of cockpitLookDeadZone is deeply into yo-yo zone. I still didn't determine where exactly (I had enough of testing for today), but it is somewhere around 0.08. Good news is that it looks like it doesn't have to be so precise, it looks like three decimal places should do. Also, on my system the fps lock at 84 is too low, regarding FFB. 169 is good. 169 is better than "no lock". "No lock" is much better than 84, and almost as good as 169. I still didn't have time to see how low can I go regarding this, but 84 is clearly too low. -------------------------------------------------------------------------------- Ok, I tested all this. First, FFB has to be precise to all 6 decimal places. My setting is 0.086551 (the default is 0.050000). With finetuned FFB I could feel better what the right FOV is, and now the best is 63. This means that the (horizontal) distance from my eyes to the edge of table where wheel is attached is 40 cm. The FOV is extremly important, with wrong FOV you cannot negotiate corners correctly. Too bad you cannot adjust it more precisely (one decimal place should be enough). Regarding what fps, the more, the better. "No lock" fps is better than 169. I tested with Skippy on Jefferson Reverse, which is a good combination for testing. |
#13
|
|||
|
|||
New PC's FFB
Mario Petrinovic:
Mario Petrinovic: Mario Petrinovic: Well, I got new PC, and there are some changes. On this PC I can clearly feel where I am regarding latency (cockpitLookDeadZone setting). Here, my latency definitelly is somewhere around 0.038500. Well, this has more sense than what I wrote previously. Thankfully now I can clearly distinguish where I am with my latency, so I can experiment with different settings (if I find time to do that). Regarding the feel of FFB, if latency is too low, your are chasing car, if it is too high, car is chasing you. When it is too low, car runs away from you, but you cannot catch it. This isn't so easy to distinguish, but it is very easy to distinguish too high latency. In short, with too high latency FFB behaves somehow like a yo-yo. Like it has a spring in itself, and this spring acts exactly like a yo-yo. It is controlable, but of course, because of latency it plays around, and in adjsuting to this, all feels just like a yo-yo. Well, now that I am sure what is what, now I can freely experiment with different video settings, to see how they affect latency. ---------------------- There is some development, here. Up till now I chased the divide between, yo-yo zone and no yo-yo zone. Well, this was wrong. On my system that divide is at cockpitLookDeadZone = 0.038570/0.038569. Well, the prefered value of cockpitLookDeadZone is deeply into yo-yo zone. I still didn't determine where exactly (I had enough of testing for today), but it is somewhere around 0.08. Good news is that it looks like it doesn't have to be so precise, it looks like three decimal places should do. Also, on my system the fps lock at 84 is too low, regarding FFB. 169 is good. 169 is better than "no lock". "No lock" is much better than 84, and almost as good as 169. I still didn't have time to see how low can I go regarding this, but 84 is clearly too low. -------------------------------------------------------------------------------- Ok, I tested all this. First, FFB has to be precise to all 6 decimal places. My setting is 0.086551 (the default is 0.050000). With finetuned FFB I could feel better what the right FOV is, and now the best is 63. This means that the (horizontal) distance from my eyes to the edge of table where wheel is attached is 40 cm. The FOV is extremly important, with wrong FOV you cannot negotiate corners correctly. Too bad you cannot adjust it more precisely (one decimal place should be enough). Regarding what fps, the more, the better. "No lock" fps is better than 169. I tested with Skippy on Jefferson Reverse, which is a good combination for testing. -------------------------------------------------------------------------------- Well, now that I've set up the things reasonably well, I tried to see how car behaves with cockpitLookDeadZone values higher than 0.1. Previously I didn't go higher than that since iRacing has set it to 0.05 as a default. Well, now I tried all the values. And look at that... So, now I have really well set up all the things, so, lets see how this will go. This value can be set in a range from 1.0 to 0.0. The middle of this is 0.5. And, look at that, this value works excellently, just like it should. Safe, and you can correct driving even if in the middle of spin (actually, it really behaves like a real car), and control it in 99% of situations. So, this SHOULD actually be the default setting. Why iRacing has put it at 10x lower value, who in the whole world knows. My God, will I spend whole my life clearing THE MESS iRacing is leaving behind. At least if they would leave some info, or whetever. But not, they are actually blocking any discussion about this. Unbelievable. So, it is actually all very easy, just put this value to 0.5, put first of FFB settings (Overall) to whatever you like, put the rest of them at 100% (I would also uncheck "Allow game to adjust settings"), put in-game FFB values at max., and set FOV so that you count in the distance from eyes to the edge of table as 40 cm. |
#14
|
|||
|
|||
New PC's FFB
Mario Petrinovic:
So, it is actually all very easy, just put this value to 0.5, put first of FFB settings (Overall) to whatever you like, put the rest of them at 100% (I would also uncheck "Allow game to adjust settings"), put in-game FFB values at max., and set FOV so that you count in the distance from eyes to the edge of table as 40 cm. -------------------------------------------------------------------------------- Oh, it turned out that setting the hight of screen is also very important. BTW, I would definitely uncheck "Allow game to adjust settings". This setting gives unlinearity to FFB. |
#15
|
|||
|
|||
New PC's FFB
On Thu, 3 May 2012 08:46:30 +0200, "Mario Petrinovic"
> wrote: >Mario Petrinovic: > So, it is actually all very easy, just put this value to 0.5, put >first of FFB settings (Overall) to whatever you like, put the rest of them >at 100% (I would also uncheck "Allow game to adjust settings"), put in-game >FFB values at max., and set FOV so that you count in the distance from eyes >to the edge of table as 40 cm. >-------------------------------------------------------------------------------- > > Oh, it turned out that setting the hight of screen is also very >important. > BTW, I would definitely uncheck "Allow game to adjust settings". >This setting gives unlinearity to FFB. Interesting. I'm beta-testing pCARS right now and with that game you definitely have to check that option. I'm surprised other games don't have the same requirement. |
#16
|
|||
|
|||
New PC's FFB
FolkGT:
Mario Petrinovic: >Mario Petrinovic: > So, it is actually all very easy, just put this value to 0.5, put >first of FFB settings (Overall) to whatever you like, put the rest of them >at 100% (I would also uncheck "Allow game to adjust settings"), put in-game >FFB values at max., and set FOV so that you count in the distance from eyes >to the edge of table as 40 cm. >-------------------------------------------------------------------------------- > > Oh, it turned out that setting the hight of screen is also very >important. > BTW, I would definitely uncheck "Allow game to adjust settings". >This setting gives unlinearity to FFB. Interesting. I'm beta-testing pCARS right now and with that game you definitely have to check that option. I'm surprised other games don't have the same requirement. -------------------------------------------------------------------------------- Are you having Damper, Spring and Centering Spring (Centering Spring unchecked) at 100%? Well, if you don't have, then some strange things happen (which are also unlinear), so, in that case, if game developer expects you to have unlinear FFB settings, he can try to flatten this unlinearity by adjusting settings (also in ulinear way). This is the case with iRacing, who is suggesting some non-linear FFB settings. |
#17
|
|||
|
|||
New PC's FFB
Mario Petrinovic:
Mario Petrinovic: So, it is actually all very easy, just put this value to 0.5, put first of FFB settings (Overall) to whatever you like, put the rest of them at 100% (I would also uncheck "Allow game to adjust settings"), put in-game FFB values at max., and set FOV so that you count in the distance from eyes to the edge of table as 40 cm. -------------------------------------------------------------------------------- Oh, it turned out that setting the hight of screen is also very important. BTW, I would definitely uncheck "Allow game to adjust settings". This setting gives unlinearity to FFB. -------------------------------------------------------------------------------- Ok, there are some corrections. Now that I've set up hight of screen, it turned out that FOV was wrong. The correct one is 66 (was 63). Previously, 63 allowed me to go through corners better, but it turned out that it was because lower FOV losenes car, and this is beneficial with default tight setups. Now, with hight of screen set correctly, I realize that 66 is the right FOV. This means that the distance from eyes to the edge of table is 36 cm, not 40 cm. |
#18
|
|||
|
|||
New PC's FFB
Mario Petrinovic:
Mario Petrinovic: Mario Petrinovic: So, it is actually all very easy, just put this value to 0.5, put first of FFB settings (Overall) to whatever you like, put the rest of them at 100% (I would also uncheck "Allow game to adjust settings"), put in-game FFB values at max., and set FOV so that you count in the distance from eyes to the edge of table as 40 cm. -------------------------------------------------------------------------------- Oh, it turned out that setting the hight of screen is also very important. BTW, I would definitely uncheck "Allow game to adjust settings". This setting gives unlinearity to FFB. -------------------------------------------------------------------------------- Ok, there are some corrections. Now that I've set up hight of screen, it turned out that FOV was wrong. The correct one is 66 (was 63). Previously, 63 allowed me to go through corners better, but it turned out that it was because lower FOV losenes car, and this is beneficial with default tight setups. Now, with hight of screen set correctly, I realize that 66 is the right FOV. This means that the distance from eyes to the edge of table is 36 cm, not 40 cm. -------------------------------------------------------------------------------- Ha, ha, I just figured out one important thing. Well, with my new PC I decided to start from the begining, from "the first state", to not change anything, to see how it will go. And immediatelly I noticed far crispier FFB. Well, I thought, that's ok, why not, : ). But I didn't know why. I ran Profiler before, I am running it now, everything was the same. Except that I noticed that I cannot change the properties of wheel axes. This is strange, I thougt. Well, today I realized that this is because I didn't have any profile loaded/selected into the Profiler. So, today I decided to make one profile, and look at that, immediatelly I noticed how much lag a profile is adding to FFB. A lot of lag, car becames almost "uncechable". Well, why don't try to turn Profiler completly off? So I did this, and guess what, things started to be even better, car became even more "catchable" (it means, when car starts to drift away, I can catch it, Watkins Glen is good track to test this, because there car slides a lot under power) than before when only Profiler without a profile was running. Well, this certainly is a good news. So, no Profiler from now on. |
#19
|
|||
|
|||
New PC's FFB
On Fri, 4 May 2012 07:32:55 +0200, "Mario Petrinovic"
> wrote: >FolkGT: >Mario Petrinovic: >>Mario Petrinovic: >> So, it is actually all very easy, just put this value to 0.5, put >>first of FFB settings (Overall) to whatever you like, put the rest of them >>at 100% (I would also uncheck "Allow game to adjust settings"), put in-game >>FFB values at max., and set FOV so that you count in the distance from eyes >>to the edge of table as 40 cm. >>-------------------------------------------------------------------------------- >> >> Oh, it turned out that setting the hight of screen is also very >>important. >> BTW, I would definitely uncheck "Allow game to adjust settings". >>This setting gives unlinearity to FFB. > >Interesting. I'm beta-testing pCARS right now and with that game you >definitely have to check that option. I'm surprised other games don't >have the same requirement. >-------------------------------------------------------------------------------- > > Are you having Damper, Spring and Centering Spring (Centering Spring >unchecked) at 100%? Well, if you don't have, then some strange things happen >(which are also unlinear), so, in that case, if game developer expects you >to have unlinear FFB settings, he can try to flatten this unlinearity by >adjusting settings (also in ulinear way). This is the case with iRacing, who >is suggesting some non-linear FFB settings. I see. The devs at pCARS recommend the following settings: Overall Effects Strength: 100-104 Spring Effect Strength: 0 Damper Effect Strength: 0 Enable Centering Spring: Unchecked. (so it doesn't matter the setting) Degrees of Rotation: 900 (different steering setups are done per car within the game) Allow game to adjust settings: Checked |
#20
|
|||
|
|||
New PC's FFB
FolkGT:
Mario Petrinovic: >FolkGT: >Mario Petrinovic: >>Mario Petrinovic: >> So, it is actually all very easy, just put this value to 0.5, put >>first of FFB settings (Overall) to whatever you like, put the rest of them >>at 100% (I would also uncheck "Allow game to adjust settings"), put >>in-game >>FFB values at max., and set FOV so that you count in the distance from >>eyes >>to the edge of table as 40 cm. >>-------------------------------------------------------------------------------- >> >> Oh, it turned out that setting the hight of screen is also very >>important. >> BTW, I would definitely uncheck "Allow game to adjust settings". >>This setting gives unlinearity to FFB. > >Interesting. I'm beta-testing pCARS right now and with that game you >definitely have to check that option. I'm surprised other games don't >have the same requirement. >-------------------------------------------------------------------------------- > > Are you having Damper, Spring and Centering Spring (Centering > Spring >unchecked) at 100%? Well, if you don't have, then some strange things >happen >(which are also unlinear), so, in that case, if game developer expects you >to have unlinear FFB settings, he can try to flatten this unlinearity by >adjusting settings (also in ulinear way). This is the case with iRacing, >who >is suggesting some non-linear FFB settings. I see. The devs at pCARS recommend the following settings: Overall Effects Strength: 100-104 Spring Effect Strength: 0 Damper Effect Strength: 0 Enable Centering Spring: Unchecked. (so it doesn't matter the setting) Degrees of Rotation: 900 (different steering setups are done per car within the game) Allow game to adjust settings: Checked -------------------------------------------------------------------------------- Once I discussed this with Ade Allen, he made Real Feel plugin fo rFactor, and he also didn't now that, and since the developers of both, iRacing and pCARS don't mention this specifically anywhere, I presume that they don't know too. Centering Spring works even if unchecked. Centering Spring is a part of FFB, and it works so that if it is more than 100% it is harder than natural for you to turn wheel away from center, and if it is below 100% it is easier than natural (or something like that). Anyway, 100% is the natural strength of FFB forces (this goes for Damper and Spring, too). When Centering Spring is checked, it works indepemdetly of FFB. IOW, this checked Centering Spring ISN'T a part of FFB, but unchecked is a part of FFB. If you see the description of this (put mouse arrow over the text), you'll see that by checking it you "enable centering forces in force feedback games which do not have a centering force". For force feedback games that do have a centering force you don't have to check this box, and this slider adjusts this force unchecked. It really is confusing (it was for me, also), and there is nowhere an closer explaination, and the whole simcommunity is convinced (just like you) that this slider doesn't work if unchecked, but this isn't so. For example, iRacing recommends this slider to be at 100%. Which means that iRacing's FFB works on "Overall", AND "Centering Spring". |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
FFB latency | Mario Petrinovic | Simulators | 2 | March 20th 12 10:14 PM |
FFB progress | Mario Petrinovic | Simulators | 5 | February 14th 12 04:36 PM |
FFB (so far) | Mario Petrinovic | Simulators | 0 | September 25th 11 10:57 AM |
FFB | Mario Petrinovic | Simulators | 26 | September 5th 11 08:52 AM |
G25 v. ECCI - To FFB or not to FFB | [email protected] | Simulators | 1 | August 17th 07 02:35 PM |