(I dunno what's going on with my video output, but I'm done tinkering for now... so here's a potato placeholder for the proper vid)
This was my last car, and this feels remarkably familiar. Top job!
Last edited by Racon, .
That's 2 separate lag-hits, one at the back and then half a second later at the front. One of you was having a low-rate moment, I think.
New wheels are here!
The 300 is updated with smaller rims, more tyre profile, wider track and some suspension and setup tweaks to fit.
The 200 (TBO spec, new wheels & tyres but no change to track) is now submitted for initial review
Both have a few tweaks to the mesh since the last version, but mostly that is still to-do.
Develpment plan for both (simultaneously) will be:
- (Any Fixes)
- Mesh finish/smooth
- Skinable
- More interior detail
The 300 is updated with smaller rims, more tyre profile, wider track and some suspension and setup tweaks to fit.
The 200 (TBO spec, new wheels & tyres but no change to track) is now submitted for initial review
Both have a few tweaks to the mesh since the last version, but mostly that is still to-do.
Develpment plan for both (simultaneously) will be:
- (Any Fixes)
- Mesh finish/smooth
- Skinable
- More interior detail
I made a suggestion (somewhere here) to allow sets to be sent to/from insim. I'd rather see that to be honest, because it would be more flexible. You could make the system you've described above with it, but you could do lots of other things too. ...and it would probably be less work for Scawen 
Nearly there with the new wheels...
The 200 and 300, as Angel & Demon:
The 200 and 300, as Angel & Demon:
Stats at https://piranmoto.co.uk/events/111
Last edited by Racon, .
My experience is that when it's good it's no worse than before, but there are 'spiky patches' where it gets randomly laggy (ie, no particular person, just everyone gets patches of low-rate or some big spikes). It was more like that at this Friday's event than the previous one, for example, a few people mentioned it.
It's not dissimilar to the start of a DDoS though, so who knows what issues the host is fighting behind the scenes - might be that we're complaining about spikiness when we've actually been spared an outage.
It's not dissimilar to the start of a DDoS though, so who knows what issues the host is fighting behind the scenes - might be that we're complaining about spikiness when we've actually been spared an outage.
**PiranMOTO Open DD, Season 7 Round 13**
**Friday 29th** at **19:00 UTC**
Live stats at https://piranmoto.co.uk/events/110
**Friday 29th** at **19:00 UTC**
Live stats at https://piranmoto.co.uk/events/110
Last edited by Racon, .
So glad to see this back, I've really missed it! We're putting it straight back in for a DD race this Friday
I would love to see that 
I have decided that this needs to be a thing in it's own right rather than just a corner of another server, so now we have PiranMOTODunsfold (with some InSim magic to keep the scoreboards free of second laps - everything there is from a standing start)
I have decided that this needs to be a thing in it's own right rather than just a corner of another server, so now we have PiranMOTODunsfold (with some InSim magic to keep the scoreboards free of second laps - everything there is from a standing start)
Dunsfold (Top Gear Test Track!) to scale at LA1X
What other choice was there for testing the new software? (link)
I started with just chalk, but it's just not visible enough at speed unless you use a huge amount of them. To get all of it in to 2400 chalk objects and still drive fast cars around it is just not possible.
So, I've settled on concrete for grass (just imagine it's blue because they left Clarkson in charge of maintaining it). Besides, I've seen how you guys treat chalk lines, so the kerb is also there to keep you honest
I've done it as autocross for a standing start, and there are restricted areas everywhere you could gain by shortcutting - don't break too late for hammerhead or you're outta there.
The Lemon Ara is a very good reasonably-priced car to start with, with 1:45ish lap times.
(EDIT: Now available online at PiranMOTODunsfoldHotlap, times available at https://piranmoto.uk/tracks/680)
(link)I started with just chalk, but it's just not visible enough at speed unless you use a huge amount of them. To get all of it in to 2400 chalk objects and still drive fast cars around it is just not possible.
So, I've settled on concrete for grass (just imagine it's blue because they left Clarkson in charge of maintaining it). Besides, I've seen how you guys treat chalk lines, so the kerb is also there to keep you honest
I've done it as autocross for a standing start, and there are restricted areas everywhere you could gain by shortcutting - don't break too late for hammerhead or you're outta there.
The Lemon Ara is a very good reasonably-priced car to start with, with 1:45ish lap times.
(EDIT: Now available online at PiranMOTODunsfoldHotlap, times available at https://piranmoto.uk/tracks/680)
Last edited by Racon, .
Reason : Moved to own server; link to laptimes, added lyt as attachment
PiranMOTO Layout-From-Image (Win console app)
(You wait years for this kind of thing, and then two come along in the same week! 
)
LFI is a command-line windows app that processes a PNG image to detect points marked in specific 'pure' colours, and outputs an LFS layout file of those points. You can scale and offset the resulting layout, making it easy to create a perfectly scaled reference layout from an image of a real-world source.
The colours used are the 8 colours available when using only full or 0 values, ie, pure black, white, red, green, blue, yellow, cyan and magenta, and each produces a different object (3 colours of cones, 3 colours of posts, a tyre stack and a pillar). See reference image.
Full colour images such as screenshots of maps need to be cleaned of these colours before you can start marking them, this can be achieved with output levels. (If your software doesn't have that, photopea.com is a free online photoshop-like software that does: Go to Image > Adjustments > Levels, and move the output levels from 0 to 1, and 255 to 254.)
Once you've made your image, you need to calculate your scale. If you've used a screenshot of a map and left the scale markings in it, you can divide the stated width in metres divided the pixel width to get the ratio. The default is 0.0625, or 16 pixels per metre. Rounding errors mean this will be close, but it is recommended to fine-tune the ratio afterwards by comparing a known length that spans as much of the layout as possible. You can calculate the perfect ratio directly: new ratio = (old ratio / wrong distance) * correct distance.
The app is limited to 2400 control points, and will stop processing with an error if more are found (this usually means you've missed a patch of pure colour when preparing the image).
The layout is created in LA1X, named after the image, and is centred at 0,0 plus any offsets you specify.
Note: If you have any important, long, straight edges, it is recommended to rotate your source image so that this line is perfectly horizontal or vertical before you paint your control points. (LFS objects rotate in 1.6 degree steps, so you can't make perfectly straight lines at just any angle. Rotating will usually anti-alias your control points and checnge their colour slightly - you can still see them, but as they're not pure colour any more, LFI will ignore them!)
Usage: lfi test.png <scale> <offset x> <offset y>
- scale is metres-per-pixel, default 0.0625 (16px/m)
- offsets are in metres, default 0
(Tested up to 3000x3000 pixels - slow, but doesn't crash. Let me know if you find the limit and can't manage without more, I can expand the memory allowed if required.)
)LFI is a command-line windows app that processes a PNG image to detect points marked in specific 'pure' colours, and outputs an LFS layout file of those points. You can scale and offset the resulting layout, making it easy to create a perfectly scaled reference layout from an image of a real-world source.
The colours used are the 8 colours available when using only full or 0 values, ie, pure black, white, red, green, blue, yellow, cyan and magenta, and each produces a different object (3 colours of cones, 3 colours of posts, a tyre stack and a pillar). See reference image.
Full colour images such as screenshots of maps need to be cleaned of these colours before you can start marking them, this can be achieved with output levels. (If your software doesn't have that, photopea.com is a free online photoshop-like software that does: Go to Image > Adjustments > Levels, and move the output levels from 0 to 1, and 255 to 254.)
Once you've made your image, you need to calculate your scale. If you've used a screenshot of a map and left the scale markings in it, you can divide the stated width in metres divided the pixel width to get the ratio. The default is 0.0625, or 16 pixels per metre. Rounding errors mean this will be close, but it is recommended to fine-tune the ratio afterwards by comparing a known length that spans as much of the layout as possible. You can calculate the perfect ratio directly: new ratio = (old ratio / wrong distance) * correct distance.
The app is limited to 2400 control points, and will stop processing with an error if more are found (this usually means you've missed a patch of pure colour when preparing the image).
The layout is created in LA1X, named after the image, and is centred at 0,0 plus any offsets you specify.
Note: If you have any important, long, straight edges, it is recommended to rotate your source image so that this line is perfectly horizontal or vertical before you paint your control points. (LFS objects rotate in 1.6 degree steps, so you can't make perfectly straight lines at just any angle. Rotating will usually anti-alias your control points and checnge their colour slightly - you can still see them, but as they're not pure colour any more, LFI will ignore them!)
Usage: lfi test.png <scale> <offset x> <offset y>
- scale is metres-per-pixel, default 0.0625 (16px/m)
- offsets are in metres, default 0
(Tested up to 3000x3000 pixels - slow, but doesn't crash. Let me know if you find the limit and can't manage without more, I can expand the memory allowed if required.)
Thanks all for the feedback (here, discord and online).
I've been experimenting a lot over the weekend with 17, 16 and 15 inch rims, and it's clear I have been trying to get too much from one car. So, I have finally decided in favour of the split: the 300 will be as before with changes listed below, and a new 200 will be TBO-spec engine, with 16" tyres in the original placement (no arches/flare in any config) and reduced rear tyre width.
The idea being that the 200 is the late 70s/early 80s base car, and the 300 is what happened when someone got hold of one in the late 80s and gave it the works. The 200 is feeling great, but needs a few set tweaks before it's ready.
For the 300, the wheels are now 15" and flush with the flare. Visually it's working, but the grip has taken a huge hit with all that sidewall - much more than the gain made from increasing the track. I'm still working on that with tyre widths, setups and suspension tweaks. It seemed much easier to work with at 16", so if all else fails I will go back to that size.
I've been experimenting a lot over the weekend with 17, 16 and 15 inch rims, and it's clear I have been trying to get too much from one car. So, I have finally decided in favour of the split: the 300 will be as before with changes listed below, and a new 200 will be TBO-spec engine, with 16" tyres in the original placement (no arches/flare in any config) and reduced rear tyre width.
The idea being that the 200 is the late 70s/early 80s base car, and the 300 is what happened when someone got hold of one in the late 80s and gave it the works. The 200 is feeling great, but needs a few set tweaks before it's ready.
For the 300, the wheels are now 15" and flush with the flare. Visually it's working, but the grip has taken a huge hit with all that sidewall - much more than the gain made from increasing the track. I'm still working on that with tyre widths, setups and suspension tweaks. It seemed much easier to work with at 16", so if all else fails I will go back to that size.
If you can stretch to an S3, there's lots of mods just for drifting.
If not, since your last visit, we've had an update to the FXR, XRR and FZR too - you can now put any tyres on 'em and also adjust the tyre sizes in the alternate configuration, so there's 3 more cars you can drift with in S2.
If not, since your last visit, we've had an update to the FXR, XRR and FZR too - you can now put any tyres on 'em and also adjust the tyre sizes in the alternate configuration, so there's 3 more cars you can drift with in S2.
3rded - they're about 50% of the mods released today
Two more mods deleted/set-private between schedule and event, booooo
In for the Vanos is the Duplato, and in for the Tomato is... the Piran Firefly 300
In for the Vanos is the Duplato, and in for the Tomato is... the Piran Firefly 300
Last edited by Racon, .
Thanks Ibta!
Feedback consensus on discord is fairly solid in that the wheels are too big and too low-profile (you guys missed the 80s? ), and that body-kit arches are no good if the wheels aren't flush to them. Fair enough - I was trying to keep the stock arches as an option but it's too much compromise visually.
So, I'm thinking of putting the kit-arches in all the configs and moving slightly smaller wheels out to fit them flush... but I'm worried about the stability. (By that I mean it's got just the right amount, IMHO, of "roll danger" and I don't want to remove too much.)
I've also been driving this a lot at TBO spec (22/23% res), and with the wheels/handling of the 300 changing I'm now thinking of making a smaller-engined version that is TBO-spec with no/small restriction - that one could have the stock arches and the current wheel track.
I will experiment
--
I've also got (much longer term) plans for a 500 version, around GTR-spec (ref pics attached). Rolling over is a problem on slicks, so that one will be more rocket bunny style, and can have all the arches and wheel-track it can handle visually
Feedback consensus on discord is fairly solid in that the wheels are too big and too low-profile (you guys missed the 80s?
), and that body-kit arches are no good if the wheels aren't flush to them. Fair enough - I was trying to keep the stock arches as an option but it's too much compromise visually.So, I'm thinking of putting the kit-arches in all the configs and moving slightly smaller wheels out to fit them flush... but I'm worried about the stability. (By that I mean it's got just the right amount, IMHO, of "roll danger" and I don't want to remove too much.)
I've also been driving this a lot at TBO spec (22/23% res), and with the wheels/handling of the 300 changing I'm now thinking of making a smaller-engined version that is TBO-spec with no/small restriction - that one could have the stock arches and the current wheel track.
I will experiment
--
I've also got (much longer term) plans for a 500 version, around GTR-spec (ref pics attached). Rolling over is a problem on slicks, so that one will be more rocket bunny style, and can have all the arches and wheel-track it can handle visually
First pass submitted for testing!
Still plenty more to do, but there's enough there it can be driven online now. (I spend more time driving it than building it, it will be nice to have some non-AI company )
Some parts from LFS cars used as is: seats, rear mirror, gearstick
Some parts modified from LFS cars: wing mirrors, steering wheel,
Some textures from LFS cars used: leather
To-do:
- Mesh smoothing (quite a few lumpy bits to fix)
- Skinable (after mesh fix)
- More interior detail (+ better texturing)
- Better body-kit arches (more detail + better fit)
Still plenty more to do, but there's enough there it can be driven online now. (I spend more time driving it than building it, it will be nice to have some non-AI company
)Some parts from LFS cars used as is: seats, rear mirror, gearstick
Some parts modified from LFS cars: wing mirrors, steering wheel,
Some textures from LFS cars used: leather
To-do:
- Mesh smoothing (quite a few lumpy bits to fix)
- Skinable (after mesh fix)
- More interior detail (+ better texturing)
- Better body-kit arches (more detail + better fit)
**snip**
Live stats at https://piranmoto.co.uk/events/106
Live stats at https://piranmoto.co.uk/events/106
Last edited by Racon, .
Slow progress on the important things (too many things to do, not enough time to do them), but I've made a minor tweaks in the meantime. Now skinable, driver-side-swapable, less likely to stall in default sets, and more coach-like wheel hubs.
Left to do:
Engine tune (*lower RPMs)
Widen
Full interior (passenger compartment)
Full dashboard (replace temporary)
Suspension/setups tune
More exterior detail/ some smoothing
Better/smoother steering animation
Make skinnable
Tidy up texture usage
*Switchable sides
Left to do:
Engine tune (*lower RPMs)
Widen
Full interior (passenger compartment)
Full dashboard (replace temporary)
Suspension/setups tune
More exterior detail/ some smoothing
Better/smoother steering animation
Make skinnable
Tidy up texture usage
*Switchable sides
Last edited by Racon, .
Last edited by Racon, .
FGED GREDG RDFGDR GSFDG