237 replies to this topic

[Steve Paradis]

Everything was in the folder ...the problem was not there at all.

it was the flashing of the firmware...

 

I went through every possible problem, even had to download that

microsoft thing because the program refused to open.

 

so here is the result :

 

 

http://www.youtube.com/watch?v=VOs9_s_EPJY

 

 

thx everyone..this is a jewel. Will start the rebuilt of my cab this weekend.

Edited by Steve Paradis, 30 January 2013 - 12:45 PM.

[Slydog43]

it that with PinDMD1 or 2?

[open6l]

it that with PinDMD1 or 2?

thats PinDMD2

[russdx]

it doesn't look like its using any of the darker shades, might need to map the shades differently?

[aupton]

Russdx,  Is this the same version of the pinDMD 2 that's currently sold out on pindmd.com ?

 

http://virtuapin.net...&products_id=98

 

I was waiting to order direct, but if its available now then I'm sold!

[Slydog43]

Can some kind folk give me directions for setting up the PinDMD2 with Future pinball, and a specific table to test on.  Thanks

[open6l]

GO to the other PINBALL site .COM

Can some kind folk give me directions for setting up the PinDMD2 with Future pinball, and a specific table to test on.  Thanks

[russdx]

Russdx,  Is this the same version of the pinDMD 2 that's currently sold out on pindmd.com ?

 

http://virtuapin.net...&products_id=98

 

I was waiting to order direct, but if its available now then I'm sold!

 

yup

[Macro]

it doesn't look like its using any of the darker shades, might need to map the shades differently?

 

for PinDMD2 at the moment it uses shades 0,6,9,12,15 for the 5 levels in FP font generated displays (since FP calls them 0%, 40%, 60%,80% and 100%) but does a linear brightness using all 16 shades for imagelist stuff.

 

since I don't have the LED DMD yet (although was told it was sent today ) I can't do much in the way of fine tuning - really need it connected to the development machine to play around with settings like that.

 

will have to set up some test screens so I can compare the two directly (LCD vs DMD) and match shades - although of course, it will only really match shades for the DMD I am using!

[aupton]

Russdx. Thanks so much for your reply. I just ordered !!!!!!

 

Now I need to decide which DMD to purchase... The quality of the Vishay's look really nice, but that X-Pin Blue DMD looks sweet too... Decisions... Decisions...

[Steve Paradis]

it doesn't look like its using any of the darker shades, might need to map the shades differently?

 

That's what I tought...it look like it is black or white only.

At first I tought it was the firmware upgrade which I messed.. (wrong DFU...or whatever)

 

As soon as there is a solution I will be happy to have it.

 

Can different OpenGl.Dll  with different setting be created for testing?

...or maybe the parameters for the 5 color range in an external txt file that we could adjust?

Edited by Steve Paradis, 30 January 2013 - 09:29 PM.

[jasonsmith]

That's exactly what we need, a test DMD script flashing all the shades one by one. Maybe just a empty table with that added so we can test it.

[Steve Paradis]

this is fairly easy to do let me knwo what kind of test pattern you want and I will do it

[jasonsmith]

How about some square blocks at the different shades with the percentages underneath? 

[Macro]

That's exactly what we need, a test DMD script flashing all the shades one by one. Maybe just a empty table with that added so we can test it.

 

I've been adding my test bits to the DMDDisplayDemo table that came with FP in the first place - mainly using an animated imagelist (nicked from stones'n'bones) and the LedWIZ stuff.

 

I would guess that the ideal would be a font screen as suggested above, with the 5 shades across the screen and a bit of text to say which is which, and the maybe a texture which has 128 shades of grey going across the screen for the 16 colour tuning - so, 0,0,0 : 2,2,2 : 4,4,4 etc up to 254,254,254  Most art packages support some form of shaded fill across an object, so should be fairly easy to do that!

 

it will also show if the code I have so far actually works - I've not actually seen what it does myself, it's perfectly feasible that there is a bug in there somewhere!

 

I could load the shades to use from some form of ini file, and just default them to the ones above if the file does not exist

 

just done this, add a file called FP-SHADES.TXT with one hex character per shade in it, and it will load them in - so it will be in the next release, few too many untested bits to let it out as it is now ... !

 

default for V1 would be 01123 and for V2 would be 069CF  (if the file does not exist, it uses these values)

Edited by Macro, 31 January 2013 - 12:07 AM.

[russdx]

ah yes the DMD gradient is NOT linear it starts off very dark and gets light very quickly iv never really worked out how to get it linear. so when mapping shades you will probably want to use mostly lower shades and a few high shades to get the best effect.

 

so for 5 levels 0,1,3,6,15    will probably give a much linear end appearance 

current map for pinmame is 0,1,5,15

 

when i do figure out how to make it linear ill let you guys no and release a new version of the firmware or if any one has any ideas? dunno if we got any programmers out there? maybe i can explain what im doing and some one could give me some pointers?

[Macro]

Ah, hadn't realised that, may change the defaults then

 

the algorithm for brightness used for images assumes all is linear as well, may need to revisit that once I have the real DMD for testing - it may be better coding for 8 shades that are nearly linear than 16 that are not!

[hotdp]

@russdx & @Macro,

Are you guys saying that the stuff I see now and LOVE (much better than hopped) Is getting even better?

[russdx]

pinmame is configured correct

but i think we can configure future pinball stuff better (macro will be able to do this when his DMD turns up and he sees the real results )

[Steve Paradis]

great news! thnx...I can live with it for now but every improvement will be chocolate

Edited by Steve Paradis, 31 January 2013 - 04:25 PM.