4027 replies to this topic

[jpsalas]

 

 

 

 

 

 

If you have a repro for that, that would be appreciated.. I thought we got rid of these a looong time ago..

 

 

 

 

 

Just have a table open for an extended duration moving lots of parts around.

That is literally all i have been doing all 3 times it happened today

 

It does not crash out

you can save, it pops 2 of those errors each time you save

but seems to do no harm other wise, but as i dont trust a friendly error, i save and quit and relaunch

 

 

It doesn't do that on any of my 2 computers. But I get a similar error when running a table, VPX quits with a similar error, but it only happens 1 times of 100. And in the mean time I may have opened several tables, copied and pasted from them. So anything could have caused that, maybe even the NVidia drivers.

[wiesshund]

 

 

 

 

 

 

 

It doesn't do that on any of my 2 computers. But I get a similar error when running a table, VPX quits with a similar error, but it only happens 1 times of 100. And in the mean time I may have opened several tables, copied and pasted from them. So anything could have caused that, maybe even the NVidia drivers.

 

 

That would be very add, as i dont have nvidia hardware at all.

On my laptop it has no GPU, just the i5's onboard graphics, and on my desktop, it is an AMD Radeon.
Strange it would show something there pertaining to nvidia?

[Thalamus]

Pretty sure that JP meant that it doesn't need to be anything directly in the vp code but something external - like video drivers. And, well, your laptop does have a GPU, built in or not.

[wiesshund]

I'd prefer to think of it more as a windows 2d accelerator 

[toxie]

pretty sure this is a VPX issue. and maybe related to some mem corruption from some little remaining bug somewhere.  :/

[fuzzel]

rev192 is up (Release Candidate)

 

- fix memory corruption in code editor

- extend light sequence blinking patterns from size '32' to 'unlimited' (fix for Star Wars Bounty Hunter)

- update LTD4a.vbs (by wiesshund)

[fourbanks]

rev192 is up (Release Candidate)

 

- fix memory corruption in code editor

- extend light sequence blinking patterns from size '32' to 'unlimited' (fix for Star Wars Bounty Hunter)

- update LTD4a.vbs (by wiesshund)

many thanks. Tables like bounty hunter are now working correctly 

[toxie]

Please let me know if there are still other tables that make problems when loading now.

[Mrtr32]

Many thanks. Star Wars Bounty Hunter now load and work great!

 

@toxie. I will test many tables as I can today and inform you if I found other issues.

Edited by Mrtr32, 30 May 2021 - 01:47 PM.

[tartzani]

good job

[wiesshund]

rev192 is up (Release Candidate)

 

 

- extend light sequence blinking patterns from size '32' to 'unlimited' (fix for Star Wars Bounty Hunter)

 

 

I'll be back in an hour
With the table of 32,000 lights

[GInsonic]

 

rev192 is up (Release Candidate)

 

 

- extend light sequence blinking patterns from size '32' to 'unlimited' (fix for Star Wars Bounty Hunter)

 

 

I'll be back in an hour
With the table of 32,000 lights

 

 

While (true) {
   light.add();

}
 

 

[ClarkKent]

Z Option in Spinner settings

[NailBuster]

Please let me know if there are still other tables that make problems when loading now.

 

for people that use PinUP Popper and want help/test if all their tables will at least startup with no issues with 10.7.....there's a PopperAutoQA.exe in your pinup folder.  Exit Popper, simply run it and select VPX (all tables) and let it run.  It will go through/test all your vpx tables and you afterwards see if any error messages come up via the snapshots it takes after running in the testQA subfolder.   You can run this any/many times as it doens't change or touch your pinup settings(or even use recording or anything).

 

FYI.  i let it run with todays build and with my 200+vpx tables on cab they all ran/started up ok.  although i did have 6 tables that made a crash/dmp files on exit (but thats the same with 10.6) it only happens on close vpx, not an big issue....IMO,  it would be a good idea in final builds to only write the crash.txt and not 300-400mb each of the dmp files if possible.

[wiesshund]

 

Please let me know if there are still other tables that make problems when loading now.

 

IMO,  it would be a good idea in final builds to only write the crash.txt and not 300-400mb each of the dmp files if possible.

 

or at least reserve it for a -debug parameter?
Some times, depending on the crash, it may be over a gig

[toxie]

 

Please let me know if there are still other tables that make problems when loading now.

 

for people that use PinUP Popper and want help/test if all their tables will at least startup with no issues with 10.7.....there's a PopperAutoQA.exe in your pinup folder.  Exit Popper, simply run it and select VPX (all tables) and let it run.  It will go through/test all your vpx tables and you afterwards see if any error messages come up via the snapshots it takes after running in the testQA subfolder.   You can run this any/many times as it doens't change or touch your pinup settings(or even use recording or anything).

 

FYI.  i let it run with todays build and with my 200+vpx tables on cab they all ran/started up ok.  although i did have 6 tables that made a crash/dmp files on exit (but thats the same with 10.6) it only happens on close vpx, not an big issue....IMO,  it would be a good idea in final builds to only write the crash.txt and not 300-400mb each of the dmp files if possible.

 

Still, if you'd have the 6 tables that crash on exit, that would be nice to check.

[dcbenji]

Last couple days been testing WEBP in 10.7 on WIP of Indiana Jones 'raytraced edition' VPW discord server.  I have the whole table baked on Octane, and in vpx there are lots of texture mapped 4k images that are being swapped for when the flashers ignite. In 10.6 table size is 280MB with already reduced and lossy 8-bit PNGs. We have 3 other tables being built this way, and these are much more complicated lighting set ups with file sizes coming it around 380MB, lossy compression. On Indy, compressing the source renders to WEBP and launching in 10.7 reduced the file size to 140MB! Plus the compression is lossless and looks fantastic on my 4k cab   no more little 8 bit dithering dots when you look up close  ,. Even the images with alpha are ridiculously reduced in size. I don't know much about this format, apparently it supports animation — does anyone know if 10.7 WEBP support includes support for animated WEBP files? Do i just need to try to make one, apply it to an object and see what happens?

 

-benji

Edited by dcbenji, 31 May 2021 - 02:42 PM.

[toxie]

Nice! Note though that even if it does support animations, then VPX could just load the first frame..  :/

But interesting. I always thought that the benefit lossless WEBP over lossless PNG (same bit depths) is only around 1/4th of savings.

[dcbenji]

Nice! Note though that even if it does support animations, then VPX could just load the first frame..  :/

But interesting. I always thought that the benefit lossless WEBP over lossless PNG (same bit depths) is only around 1/4th of savings.

ah ok i figured, no worries. perhaps my PNG source renders are not as optimized as they could be to begin with. But even so, setting the WEBP compression to 90% is giving me huge reductions in filesize with extremely little noticable change in image quality

[DeeVPin666]

Anyone have issues with Stanger Things and 10.7.0-185? On mine I have the grey under flips on top which causes the ball to get stuck on the rail. Thanks