Racelogic Video VBOX HD2 Review
#136
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,097
Received 3,270 Likes
on
1,865 Posts
Oh, man! I've come into a treasure trove of CAN info for modern cars!
C7, Viper ACR and GenV, GT3 Cups 997.2/991.1/991.2, etc. SnB, let's catch up!
A little birdie told me this is going to be much better and much easier to access for folks this summer...
C7, Viper ACR and GenV, GT3 Cups 997.2/991.1/991.2, etc. SnB, let's catch up!
A little birdie told me this is going to be much better and much easier to access for folks this summer...
__________________
-Peter Krause
www.peterkrause.net
www.gofasternow.com
"Combining the Art and Science of Driving Fast!"
Specializing in Professional, Private Driver Performance Evaluation and Optimization
Consultation Available Remotely and at VIRginia International Raceway
-Peter Krause
www.peterkrause.net
www.gofasternow.com
"Combining the Art and Science of Driving Fast!"
Specializing in Professional, Private Driver Performance Evaluation and Optimization
Consultation Available Remotely and at VIRginia International Raceway
#137
The Penguin King
Rennlist Member
Rennlist Member
#138
Rennlist Member
Thread Starter
#139
"Bugs" in the GT4 Scene
After not being able to track in April because of other commitments, I finally got back on the horse yesterday running the GT4 scene from ShakeNBake. I wanted to point out a few "bugs" (for lack of a better word) to see if any edits have been made recently.
Either the steering position needle or the RPM needle has some sort of problem with transparency where in many conditions it will cause the RPM gauge or the needle to not fully render. I'm attaching a few screenshots where you will see the needle disappear or other parts of the gauge disappear (like the gear indicator.)
If you want to see this in real time - watching any of this view will show that error(s) manifest itself. Just watch the RPM gauge.
It's easiest to see when the steering position is far left (watch the gear number digital indicator disappears) or the steering position and the RPM needle intersect at certain angles/points (the needle disappears)
Either the steering position needle or the RPM needle has some sort of problem with transparency where in many conditions it will cause the RPM gauge or the needle to not fully render. I'm attaching a few screenshots where you will see the needle disappear or other parts of the gauge disappear (like the gear indicator.)
If you want to see this in real time - watching any of this view will show that error(s) manifest itself. Just watch the RPM gauge.
It's easiest to see when the steering position is far left (watch the gear number digital indicator disappears) or the steering position and the RPM needle intersect at certain angles/points (the needle disappears)
#140
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,097
Received 3,270 Likes
on
1,865 Posts
Easy fix within the element layering...
Looks good!
Looks good!
#141
Rennlist Member
Thread Starter
After not being able to track in April because of other commitments, I finally got back on the horse yesterday running the GT4 scene from ShakeNBake. I wanted to point out a few "bugs" (for lack of a better word) to see if any edits have been made recently.
Either the steering position needle or the RPM needle has some sort of problem with transparency where in many conditions it will cause the RPM gauge or the needle to not fully render. I'm attaching a few screenshots where you will see the needle disappear or other parts of the gauge disappear (like the gear indicator.)
If you want to see this in real time - watching any of this view will show that error(s) manifest itself. Just watch the RPM gauge.
https://www.youtube.com/watch?v=C4noI6UYevA
It's easiest to see when the steering position is far left (watch the gear number digital indicator disappears) or the steering position and the RPM needle intersect at certain angles/points (the needle disappears)
Either the steering position needle or the RPM needle has some sort of problem with transparency where in many conditions it will cause the RPM gauge or the needle to not fully render. I'm attaching a few screenshots where you will see the needle disappear or other parts of the gauge disappear (like the gear indicator.)
If you want to see this in real time - watching any of this view will show that error(s) manifest itself. Just watch the RPM gauge.
https://www.youtube.com/watch?v=C4noI6UYevA
It's easiest to see when the steering position is far left (watch the gear number digital indicator disappears) or the steering position and the RPM needle intersect at certain angles/points (the needle disappears)
#142
Indeed, when I looked at the scene, it didn't seem like an ordering issue. Everything is on top of what it should be on top of.
My best guess is that an image with transparency (needle.png) on top of another image or images with transparency is not well liked by the renderer? ...
I saved out a few different versions to test with. One with the steering needle, but not the RPM needle. One without the steering needle, but with the RPM needle.
Hopefully I'll have some time tommorow to load both and drive around my "test track" and give us enough intel to give them a proper bug report with a repro case.
Thanks as always. Still a fan of the device.
-V
#143
Rennlist Member
Thread Starter
Happy to help you develop, or just QA/test. I'm fairly technical. Thanks for the fast reply letting me know you know about it.
Indeed, when I looked at the scene, it didn't seem like an ordering issue. Everything is on top of what it should be on top of.
My best guess is that an image with transparency (needle.png) on top of another image or images with transparency is not well liked by the renderer? ...
I saved out a few different versions to test with. One with the steering needle, but not the RPM needle. One without the steering needle, but with the RPM needle.
Hopefully I'll have some time tommorow to load both and drive around my "test track" and give us enough intel to give them a proper bug report with a repro case.
Thanks as always. Still a fan of the device.
-V
Indeed, when I looked at the scene, it didn't seem like an ordering issue. Everything is on top of what it should be on top of.
My best guess is that an image with transparency (needle.png) on top of another image or images with transparency is not well liked by the renderer? ...
I saved out a few different versions to test with. One with the steering needle, but not the RPM needle. One without the steering needle, but with the RPM needle.
Hopefully I'll have some time tommorow to load both and drive around my "test track" and give us enough intel to give them a proper bug report with a repro case.
Thanks as always. Still a fan of the device.
-V
#144
Rennlist Member
Not that "me too" helps solve the problem but I can repro the same symptoms and I also believe that it's not a problem with ordering. Feels more like an alpha blend math bug but that's just intuition.
Here's an example video that shows the problem.
Watching that you can see that the steering "needle" is occluding the RPM needle with a rectangular mask area...the puzzling part is that the width of that that masking area seems much much wider than the steering needle image -- it's almost like the long skinny needle image got placed in a square layer that is rotated over the layers below...the steering needle layer correctly shows transparent where the alpha channel info says to -- and indeed the background image (gauge face) shows through -- but the other co-located moving layer, the RPM needle is just masked by the top layer steering needle surface regardless of the alpha state of any overlaid pixel from the steering needle surface.
My guess is that any overlapping moving elements of a scene that have some transparency in the image to be moved would show the same symptoms.
Here's an example video that shows the problem.
Watching that you can see that the steering "needle" is occluding the RPM needle with a rectangular mask area...the puzzling part is that the width of that that masking area seems much much wider than the steering needle image -- it's almost like the long skinny needle image got placed in a square layer that is rotated over the layers below...the steering needle layer correctly shows transparent where the alpha channel info says to -- and indeed the background image (gauge face) shows through -- but the other co-located moving layer, the RPM needle is just masked by the top layer steering needle surface regardless of the alpha state of any overlaid pixel from the steering needle surface.
My guess is that any overlapping moving elements of a scene that have some transparency in the image to be moved would show the same symptoms.
#145
Rennlist Member
Thread Starter
You can see a similar thing happening in this scene as well:
It seems to be this transparent needle concept.
It seems to be this transparent needle concept.
#146
Rennlist Member
Yes, and you can see from that one that the general case is true: the top layer that has transparency is occluding any moving/animated layers below it where there's an overlap of elements....the only thing coming through the top layer transparency properly is the background static image.
Are you already working this as a bug report with RaceLogic??
Are you already working this as a bug report with RaceLogic??
#147
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,097
Received 3,270 Likes
on
1,865 Posts
Fellows, every email or contact with support at Racelogic generates a freshdesk support ticket number. If there isn't one, then there is no open ticket.
These are both custom scenes with a variety of custom elements. While I have no doubt this can be fixed, there are other, proven elements that relate the same info in a more discrete way. By combining so many elements into one combination, issues have been observed that would not normally come to light.
There is no system that is as customizable or user tailored as VBOX, and this bug is not affecting the usability of the unit or the function. Just my .02
These are both custom scenes with a variety of custom elements. While I have no doubt this can be fixed, there are other, proven elements that relate the same info in a more discrete way. By combining so many elements into one combination, issues have been observed that would not normally come to light.
There is no system that is as customizable or user tailored as VBOX, and this bug is not affecting the usability of the unit or the function. Just my .02
#148
Rennlist Member
Of course, Peter. That's why I asked whether there's a ticket open with RL already or not -- no sense in bombing their system with the same issue over and over.
Agree there are workarounds -- I made a separate steering wheel graphic with alpha and that works...but Jacy's combined elements is an intuitive and efficient use of screen area. The fact that CT supports transparency for animated elements at all implies that layering them isn't an unreasonable expectation but as a s/w developer by trade I can easily see how that might not have been tested by RL's QA folks.
I'll take your word for it on the first part of that...you've seen way more than I. Agree also that the bug isn't anything like a showstopper and workarounds are easy. I hope you will agree that it still makes sense to report the issue and let RL respond, with luck by fixing it, rather than just giving up and using discretely tiled elements alone and not reporting the problem.
Agree there are workarounds -- I made a separate steering wheel graphic with alpha and that works...but Jacy's combined elements is an intuitive and efficient use of screen area. The fact that CT supports transparency for animated elements at all implies that layering them isn't an unreasonable expectation but as a s/w developer by trade I can easily see how that might not have been tested by RL's QA folks.
There is no system that is as customizable or user tailored as VBOX, and this bug is not affecting the usability of the unit or the function.
#149
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,097
Received 3,270 Likes
on
1,865 Posts
I've been impressed with AiM Sports and Racelogic USA's follow up on reported issues.
#150
Rennlist Member
Thread Starter
I sent in the issue today, we'll see how it goes. It's possible this is a function of a transparent needle - like what I'm using for the steering marker. Would help force the prioritization if a couple other folks also sent it in as well.
This does not happen on the VBL - i use the same element configuration and graphics for the steering marker.
This does not happen on the VBL - i use the same element configuration and graphics for the steering marker.