Consulting DarkTimes' expertise once more concerning this packet:

• In single player intersecting Autocross objects can't be added. Nor can objects be added out of bounds. Is there a workaround? I assume there isn't but I would rather be sure.
• I'm having a hard time making use of the axm.UCID. I'm trying to find out who is adding/deleting a specific object, unsuccessfully so far.

In light of what happened today

What do you mean exactly? You can always save your layout with shift + U. Nice layouts btw, I like the speedhump/tyres portion and your "mothership"

They are currently on WS • Metropolis although I change the layouts every now and then. Feel free to change them but warn those on the ramps first though

It might be fun to have the ability to change the altitude of objects even above the track but maybe it is intentional. Perhaps the developers don't want this to be possible but that's pure speculation.

A few screenshots of another little test:

Quote :

How i can set the Height? All what i try i cant make my Ramp up in the air

You can only modify the altitude of out of bounds objects (Altitude is the parameter in blue) so you can't put objects in the air above the track.

FYI: Upcomming version of ClaViCo AP will also work in single player mode.

I recall Scawen mentioning the possibility to post the dimensions of autocross objects a few months ago, around the release of 0.6B if my memory serves me right. Was this information made available in the end?

It would help me alot while I'm working on a new version of my autocross tool ClaViCo AP (http://www.lfsforum.net/showthread.php?t=77749) and I'm sure I'm not the only one.

You can measure the autocross objects with the autocross measuring tool (default key D) but it remains a rather approximative and non-reliable way to get the dimensions of autocross objects.

Furthermore does anyone recall which is the minimum step autocross can be moved by. Was it 25cm? Again I couldn't find the post wherein Scawen mentioned it. Thanks in advance.

sarxes, I will try to write a more detailed readme file when I release the new version of ClaViCo AP.

"Skyroads" can be made with the following values:

W:44
L:155
A:3
LU

I'm glad some LFSers are still having fun with this tool. A new version of the tool is nearing completion. It will remain Alpha (due to lack of testing and to give me an excuse to go wild feature wise if I feel the need to) but I'm confident it will be stable. Unfortunately I have no idea when I can/will release it. The following have been implemented/fixed/changed in my version but do not take this for granted, it ban be subjected to changes at any time:

changes:
=======
• Object limit raised from 50 to 300
• Object log system removed
• Fix of numerous small bugs
• Optimisation

new features:
==========
• custom patterns (ability to setup object patterns)
• quick removal objects (10)
• ClaViCo AP can now be used in combination with the LFS Autocross Editor (NOTE: editors overlap each others)
• save/load/remove all sublayouts
• save/load/remove trackpaths

ClaViCo AP is the most interesting Plugin featured in ClaViCo but I'm also working on a few others (those are not shared publicly). Therefore I can't give an ETA, not even an approximation. Thanks for being patient though.

Interesting, I think there is an accurate way to be found combining the data in the IS_CON packet and the IS_MCI packet. At the moment I'm trying to find a way to calculate if car A was driving towards car B while not showing a sign of slowing down if the magnitude of the impact is greater than a constant value.

I'm currently fiddling with the IS_CON packet. I wondered if someone has found an appropriate way to distinguish an intentional crash from an accidental crash. Any advise about this matter?

The unjustified and destructive posts of some forum members should really stop. All this negativity is unfair to the developers and unfair to the LFS fans.

It may well be frustrating being "obliged" to wait for a long awaited update but venting your frustration in the most outragious possible way on the forum will not really help, not even slightly...
The devs have always made pertinent progress reports. Scawen in particular is doing everything he possibly can to make good progress. As Victor pointed out, the challenge Scawen is facing is not to be taken lightly.

It is rather childish to pressure him to release unfinished things. The last thing you want is for an update to be rushed through, resulting in a mess some software developers get into. In my opinion Scawen is a perfectionist and thank God he is. LFS wouldn't be so good if it wasn't for that continious pursuit of perfection.

I also find it unjust and immoral to reduce the LFS developers to "PatchMachines". They are working hard and with passion to improve LFS further and in return they get they get the "megatrolluberlolinyourfacewearealldoomed" treatment. It doesn't feel right which is why I posted this rant.

Personally, I have always been pleasantly surprised by LFS updates. Never has an LFS patch disapointed me and I still adore 0.6B to bits. As long as the LFS devs are "on the ball" I will remain supportive.

This is a rather compelling database "model". I might make it compliant with my application

Quote :

Edit: I realised now that you might be trying to use this on a dedi-server, which won't work. Switching to SHIFT+U mode only works if you're connected to a local copy of LFS.

I knew I overlooked something

Did any of you guys experience stability issues with VS 11 beta (ultimate)? I had 3 rather random crashes after roughly 2 hours of use. Apart from these I must say I like VS11.

I applied a "hybrid" solution concerning the IS_AXM receive events. I have alot more to test though, for example is their a significant performance difference if you don't set the ISF_AXM_LOAD flag etc.

IS_CPP remains a mystery to me. Perhaps I fail to setup the viewPLID correctly. Can you confirm you tested it successfully before?

Hello,

I came across an interesting interview about tyre physics which may help you to understand the difficulties and challenges developers face when writing their simulator. If you watched the whole video you might become more patient in relation to LFS development

Personally, it gave me an insight on the size of the challenge Scawen is currently facing. Beware, this video can be rather technical so it might not be your cup of tea if you are not interested in knowing more about tyre physics. This interview made me rather curious, I wonder which road Scawen choose or how LFS tyre physics relate to what is discussed during the interview. Hopefully, some of you are "knowledge hungry" and will enjoy this video. The interview ends with an optimistic conclusion. Thanks to the evolution of hardware developers now have more room for improvements. Some may also understand why it takes Scawen time to develop tyre physics, it is by all means not an easy challenge ;-)

http://channel9.msdn.com/blogs ... time-simulation-explained

IS_AXM
======
Thanks for the suggestions. Testing with a Stopwatch gave me values well under 1 MilliSecond (between 100 to 1000 stopwatch ticks but I don't know how many NanoSeconds/MicroSeconds 1 stopwatch tick is) both for Add & Del. The issue must be elsewhere.

Is it a certainty that receiving IS_AXM packets does not take significantly more time than sending IS_AXM packets? There is a high probability that I'm doing something else wrong. Furthermore I noticed that I didn't setup the UCID when sending the IS_AXM packet. I haven't found out how to setup the UCID correctly yet. More testing will have to be done before I can narrow it down. At the moment it just seems faster to Add & Del objects repeatedly without using the IS_AXM receive event.

My explanations may be incomprehensible. You would understand the problem on sight if you try it out yourself on my server. If you want to we could meet for a few minutes on my server?

IS_CPP
=====
Concerning other packets, do you know an efficient way to force the shift+U view with the IS_CPP packet? I'm currently fiddling with it unsuccessfully.

Last edited by sicotange, Thu 8 Mar 2012, 21:52. Reason : removed IS_STA bug report, no issues here

Quote :

There was a small bug that crept into the code for v2.0.14 that I have now fixed.

I confirm, a good thing it was an easy fix

Quote :

What sort of delay are we talking about?

It's a small almost irrelevant delay. Around 100ms to make a wild guess.

Quote :

Is it a delay between you sending the request to delete the packet and the packet being deleted? Or is it a delay between when you delete an object in shift+u mode and the packet being received?

The delay occurs when you send multiple IS_AXM Del requests one after another. The first Del request seems barely executed when the next one is sent. That is where the delay is noticable. It only happens when you spam Del requests (not using shift+U). I suspect my code might be to blame, it currently stores the Object Info in a List. The delay might be caused when a request is made to retrieve the stored Object Info (in order to delete the object) from the List.
And the delay is barely (if at all) noticable when you select and delete an object with shift + U using your mouse.

Quote :

Also are connected to a local copy of LFS when you notice the delay, or to a remote dedicated server?

I only tested locally but will test it soon online. Hopefully I'm to blame. The process of storing, reading and retrieving Object Information from a List multiple times per second could be the cause?

I have been using InSim.NET for over a year now and I'm very happy with it I'm glad you are still on the ball and continue improving this library. Over the past few months I used the 2.0.12 version. I recently switched to 2.0.13 and everything seemed to work well. No major anomaly whatsoever (apart from a little concern I will mention below). 2.0.14 is a different story unfortunately...

concerning 2.0.14:
=============

2.0.14 hasn't been "plug & play" so far. For some reason LFS dedi shouts "InSim: password does not match your multiplayer admin password". I have not tested in debt yet but it is the first time I experience compatibility issues. All previous versions didn't require me to alter any of my codes. Could it be IS_ISI related? I will pursue my investigation and report back once I have more solid information.

concerning 2.0.12 & 2.0.13:
===================

I have spent some time fiddling with the IS_AXM packet for the past few months. When a packet is received with the ActionFlags.PMO_DEL_OBJECTS flag I noticed a small but noticable delay. It only occurs with this parameter. The delay didn't seem to be influenced by the amount of Object Information included in the packet. I have not yet found out a good way to benchmark this and am not 100% convinced it is related to the library. Again the delay is only noticable when you delete objects. It works flawlessly when you add objects.

Last edited by sicotange, Tue 6 Mar 2012, 21:05. Reason : present perfect of "to spend" is "spent" not "spend", shame on you sicotange!

For those still interested in this tool, I have made plenty of progress for the upcomming version. The update is still far away I'm afraid but I wish to release it in time for summer. In the meantime you may like the random layout I made recently. It took me just under 5 minutes to go from the beginning till the end in an FZ5

A quick update concerning ClaViCo AP.

Ramp2 width
=========
I'm in the midst of a mini dilemma: the Ramp2 width. Scawen mentioned that the Ramp2 object is 2,8m wide. The current version of ClaViCo AP uses a width of 2,75m. Using a value of 2,8125m is also possible. I'm wondering which parameter gives the best result in terms of allignement.

Ramp2 "twins"
==========
One of the new features I have been working on is a small pattern which automatically adds 2 Ramp2 objects alligned in width ("twins").

Object logs
=========
Object logs can now be disabled and enabled at any time. One little issue surfaced so it still requires polishing.

I won't give an ETA, I can only say that progress is being made little by little

I'm afraid I will have to delay the scheduled release. I found a new way to optimise a few things and wish to do that first. It's always a risky thing to post a release schedule knowing very well that unforeseen situations can subsist. I will release the new version when it's ready and hope to reach a Beta status when the new LFS patch is released

In the upcomming version of ClaViCo AP you will be able to enable/disable object logs. When disabled there won't be any limit anymore. I plan to release the next version around next Friday.

Quote :

Yes - add one ramp,position where you want,add 2nd beside,adjust the position you need etc. Quite time consuming thing,but you see the results alredy in other posts.

You do realise you can make use of the Indexers? These indexers are used to increment or decrement the value in question (W:, L:, A:, H: ) so each time you click the "Add" button the next object will be added incremented/decremented by the values of the indexers. They are very usefull to quickly make patterns like ovals, bankings etc.

EDIT:

For example:

To make the pattern you see in the screenshot the following Indexers values were used W:20 L:155 A:0 H:3. Once I setup the Indexers I did have to click "Add" frantically 85 times though (the pattern you see is made of 85 Ramp2 objects). Next release might include a faster way to make your pattern so you wont have to click "Add" frantically.

Last edited by sicotange, Fri 13 Jan 2012, 23:01. Reason : added example

Quote :

I don't seem to be able to add distance Markers, if I try to, for example, add a 100m sign it does a red barrier instead, Markers for corners do work.

Thanks for that bug report, it has been fixed in my version. This bug won't be included in next release

You are using it in conjunction with LFS dedi? If so perhaps post your LFS dedi setup.cfg settings.

One more thing, I tested ClaViCo AP on the following operating systems successfully:

• Windows7 32bits
• Windows7 64bits
• Windows8 Developers Preview 64bits

Are you using XP or Vista?