|
Post by Casethecorvetteman on Jul 19, 2015 4:42:50 GMT -5
Hi all I need to line double the 15.7kHz 240p ( it is indeed a 240p image and not 480i before anyone asks ) output of my video game consoles and arcade boards to 31.5kHz to use these on the Cine 9, the internal LIMO-PRO scaler does this fine, but what I want to know is how much lag there will be if set to line double? Essentially all it looks to be doing is repeating each line a 2nd time, which I would reckon it should be able to do on the fly with no frame lag at all, but I don't know for sure. Every second line looks 100% identical to the line above it on that mode. What I would rather, is if there is some way to retain the exact 240p image on screen, and every 2nd line be black, so the image would then appear as it should, instead of jaggy from the doubling. Im not sure if anything can do this. What I really NEED though, is something that will change 15.7kHz to 31.5kHz with no frame lag at all. Just one frame of lag will prevent me from doing what I want to do. Thanks in advance
|
|
|
Post by gjaky on Jul 20, 2015 2:13:48 GMT -5
Absolute zero lag is not possible IMHO. The simpliest method would be indeed to simply duplicate each line, but that only could happen in the digital domain: analog signal -> A/D -> Memory -> D/A -> doubled line This has to be done in synchronized frequency to the original signal, so the first few frames in the video stream-session would eventually lost during syncing, but that can be neglected I think, since that is usually just a logo on these machines. With this method and with relatively cheap electronics a few lines of lag might be possible, with expensive (compared to the market value of this kind of device) the lag might remain within a line.
|
|
|
Post by Casethecorvetteman on Jul 20, 2015 3:52:02 GMT -5
Its got to remain very near within a couple lines at worst, any more than 5 or so lines and it is too much lag to work.
|
|
|
Post by barclay66 on Jul 20, 2015 5:27:30 GMT -5
Every second line looks 100% identical to the line above it on that mode. What I would rather, is if there is some way to retain the exact 240p image on screen, and every 2nd line be black, so the image would then appear as it should, instead of jaggy from the doubling. Im not sure if anything can do this. Hi, You might be able to acheive this if a specific signal can be found somewhere in the line doubler circuit: The signal that indicates whether an even or an odd line is being outputted. Using this signal You could blank out either all even or odd lines. Do You have any schematics of the BR909 line doubler? Regards, barclay66
|
|
|
Post by Casethecorvetteman on Jul 20, 2015 19:06:22 GMT -5
Not that i know of, but will doing this do anything for lag?
The issue is game guns, when the trigger is pulled, the screen flashes white for one frame, which the gun sees, and reports when it sees the white, thus knowing where youre pointing. With a frame of lag, it will assume youre pointing off sceen.
|
|
|
Post by mastertech on Jul 20, 2015 20:48:34 GMT -5
Hmmm, another schematic not in that service manual. If anyone has it and can send it to me i will see it gets posted.
|
|
|
Post by barclay66 on Jul 21, 2015 9:12:57 GMT -5
Not that i know of, but will doing this do anything for lag? Well, If the LimoPro is a real Line Doubler there shouldn't be any (if it does its job on a per-line level) or a maximum of one frame lag (the time it needs for digitizing an incoming frame and outputting it at double speed by duplicating each line). This is something that You will have to find out by testing. It could be measured using a two-channel oscilloscope where You feed the input signal to channel 1 (shifted to the top of the scope's screen) and the LimoPro's output to channel 2 (shifted to the bottom of the scope's screen). If the two curves have their peaks and lows at the same horizontal position then there's no lag. If channel 2 is shifted to the right, then the lag can be calculated by the distance of the shift multiplied with the timebase setting (ms/div or µs/div). I might be able to do that with the VP30 and the VP50Pro that I have... Regards, barclay66
|
|
|
Post by Casethecorvetteman on Jul 21, 2015 21:43:10 GMT -5
The only way i can test is to set up the arcade boards and test with a light gun, which i havent had time to do yet.
Id be interested to know though if there are other options that will work, id expect up to 5 lines of lag shouldnt matter too much when the gun can be calibrated.
|
|
|
Post by barclay66 on Jul 29, 2015 2:32:52 GMT -5
|
|
|
Post by Casethecorvetteman on Aug 2, 2015 3:36:27 GMT -5
I tested the Namco Guncons today via the LIMO-PRO and they were not working properly, when they did register a shot it was nowhere near where the gun was pointed, so the lag is too high.
|
|
|
Post by Casethecorvetteman on Aug 2, 2015 3:56:50 GMT -5
Those do look quite interesting Thanks!!
|
|
|
Post by Casethecorvetteman on Aug 4, 2015 21:27:20 GMT -5
Ok, ive bought an XRGB3 because they claim a 2ms lag when line doubling. Im not sure how many scan lines that will be at 60Hz, but should be ok.
|
|
|
Post by gjaky on Aug 5, 2015 0:26:56 GMT -5
That is about 1/3 line lag.
|
|
|
Post by barclay66 on Aug 5, 2015 2:27:02 GMT -5
Hi,
I don't think that it is 1/3 of a line only.
At 60Hz we have a complete frame every 16,667ms (1/f=T: 1/60=0,01667s=16,667ms). So if we have a lag of 2ms that will be roughly 12% of a frame (2ms/16,667ms*100=11,999). At 240 lines per frame this will result in a lag of about 29 lines (240*12/100=28,8)...
Regards, barclay66
|
|
|
Post by gjaky on Aug 5, 2015 5:19:34 GMT -5
Hi, I don't think that it is 1/3 of a line only. At 60Hz we have a complete frame every 16,667ms (1/f=T: 1/60=0,01667s=16,667ms). So if we have a lag of 2ms that will be roughly 12% of a frame (2ms/16,667ms*100=11,999). At 240 lines per frame this will result in a lag of about 29 lines (240*12/100=28,8)... Regards, barclay66 Indeed.. my bad.
|
|