[Coco] Last video I promise... :)

Fri Dec 5 12:27:41 EST 2014

Luis, I'm sorry to hear that you are considering/intending to remove the
artifact ability.

I've not received my "surplus" funds to purchase the needed pieces yet to
build one myself; but, would a shielded cable to bypass the possible noisy
traces help?

Shsin
On Dec 5, 2014 11:09 AM, "Luis Antoniosi (CoCoDemus)" <
retrocanada76 at gmail.com> wrote:

> Ok finally my mules decided to arrive and I assembled one for me.
>
> I don't get the noise you have there, but I noticed there is a small noise.
> My prototype doesn't have this. The culprit is the VGA ouput. In my
> prototype I have placed them far away from the input lines. But in the new
> design i tried to make things more compact and I sucked on this :P
>
> Well, modifying the FPGA I have reduced the amout of noise changing the
> input/output currents. You can check whether your noise is the vga by
> doing:
>
> screen 80
> cls 4
> dir
>
> Check if the noise increases as the amount of the text on screen also
> increases. In your video, when you blast the trimmers, I can clearly see
> blanking lines waving, so it must be it.
>
> How to solve this completely ? Redesign the board ? Using a faraday cage ?
>
> Update your repo and try the new version. it might help.
>
> Regarding the colors, you may notice that some colors have a "noise
> pattern" on that, look that this partern doesn't run, it has some
> flickering but they stay put.
>
> This is not external interference, it's merely slack time that is affecting
> how the colors are digitized. Let's say: the altera is running the code at
> the limit of it's speed. This cause the code to get out of sync with the
> clock and then, the color is digitized in the wrong level of the DAC, or it
> didn't have time enough for the differential to change its value. Solving
> this is much more difficult.
>
> The first thing is to remove and simplify the code, this means no more
> artifacting colors, and other stuff. This would help getting a stable
> timing though.
>
>
>
>
>
>
>
> Luis Felipe Antoniosi
>
>
>
> On Fri, Nov 28, 2014 at 11:11 PM, Richard Goedeken <
> Richard at fascinationsoftware.com> wrote:
>
> > Hi Luis,
> >
> > Here are new images of the board and the video output:
> >
> > http://www.fascinationsoftware.com/media/rgb2vga/IMG_9763.JPG
> > http://www.fascinationsoftware.com/media/rgb2vga/IMG_9766.JPG
> > http://www.fascinationsoftware.com/media/rgb2vga/IMG_9768.JPG
> > http://www.fascinationsoftware.com/media/rgb2vga/IMG_9770.JPG
> > http://www.fascinationsoftware.com/media/rgb2vga/IMG_9775.JPG
> >
> > Unfortunately, grounding the connector shield didn't fix the noise issue;
> > it's
> > still there like before.  I'll have to consider all of the various
> devices
> > that are plugged in, and see if there is a ground loop there somewhere.
> I
> > am
> > using 1% resistors for the 75 ohms.  Do you have any suggestions for
> > improving
> > the color fidelity?  What changes did you make to the FPGA code to
> > customize
> > it for your particular Coco?
> >
> > Richard
> >
> > On 11/28/2014 05:55 PM, Luis Antoniosi (CoCoDemus) wrote:
> > > now I saw the video.
> > >
> > > First question: which program is that ? How do I get it. It's very nice
> > for
> > > testing the colors.
> > >
> > > Second, can you take a picture of your board ? bottom/top ?
> > >
> > > There is indeed a noise band rolling. I don't get these, but my board
> > from
> > > osh park DIDN"T ARRIVE YET. I takes a month to cross the canadian
> border.
> > > Dude they must put them on the back of mules. Cannot explain that.
> > >
> > > Then check the grounding. Make sure everything is grounded from coco,
> > > altera and the monitor. (make sure the outer shell is grounded as I
> > said).
> > > It's getting there, the image is pretty good already.
> > >
> > > Also, do you have MPI or any other device  on your coco ? I'm always
> > > testing with either superIDE or CocoSDC but never mpi. Just try the
> coco
> > > without any device and check the 32 and 80 columns screens:
> > >
> > > width 80
> > > 10 for n=32 to 255
> > > 20 print chr$(n);
> > > 40 next n
> > >
> > >
> > >
> > >
> > > Luis Felipe Antoniosi
> > >
> > >
> > >
> > > On Fri, Nov 28, 2014 at 8:02 PM, Richard Goedeken <
> > > Richard at fascinationsoftware.com> wrote:
> > >
> > >> Hi Luis,
> > >>
> > >> Thank you so much for designing the awesome video converter and making
> > all
> > >> the
> > >> plans available.  I built one up yesterday and hooked it up to my Coco
> > 3,
> > >> and
> > >> it works, but there is some noise in the video output and the colors
> are
> > >> not
> > >> exactly right.  Can you help me get it working?
> > >>
> > >> I uploaded a video which shows the problems very well.  It's raw
> mpeg2ts
> > >> (AVCHD) at 1080i 60hz from my video camera; the file is about 131MB.
> I
> > >> didn't
> > >> want to upload it to youtube which would crush the resolution and
> frame
> > >> rate.
> > >>  You should be able to view it with VLC or mplayer or similar:
> > >>
> > >> http://www.fascinationsoftware.com/media/Coco3-RGBVGA.mts
> > >>
> > >> In the video, I turn the 10k potentiometer up and down to show how it
> > >> affects
> > >> the colors.  You can see that there are some rolling bands of noise in
> > the
> > >> brightness values of the pixels.  And many of the colors seem to be
> the
> > >> same
> > >> shade.
> > >>
> > >> The test screen output (when the Coco is turned off) is perfect, so
> the
> > >> problem is with the input section.  I wasn't sure exactly which build
> of
> > >> your
> > >> FPGA code I should use.  And I've never used FPGA tools before, so
> there
> > >> was a
> > >> little bit of a learning curve in installing the tools and using them.
> > >> But I
> > >> ended up with the latest git clone from the github repo (as of
> > yesterday),
> > >> and
> > >> I used the Quartus programmer tool to write the output_file.jic into
> the
> > >> serial eeprom loader memory.  Should I have used a different version?
> > Do I
> > >> need to compile the fpga code?  It looks from your repo that you
> update
> > the
> > >> SOF and JIC files with every commit, so I assume that means I don't
> > need to
> > >> compile the project if I don't make any of my own changes.
> > >>
> > >> Any ways, thanks again for this great product, and I hope I can get it
> > >> working
> > >> here.
> > >>
> > >> Richard
> > >>
> > >> On 11/11/2014 05:03 PM, Luis Antoniosi (CoCoDemus) wrote:
> > >>> It's all here:
> > >>>
> > >>> https://sites.google.com/site/tandycocoloco/rgb2vga
> > >>>
> > >>>
> > >>> Luis Felipe Antoniosi
> > >>>
> > >>>
> > >>>
> > >>> On Tue, Nov 11, 2014 at 7:23 PM, Mark McDougall <
> msmcdoug at iinet.net.au
> > >
> > >>> wrote:
> > >>>
> > >>>> On 12/11/2014 2:42 AM, Luis Antoniosi (CoCoDemus) wrote:
> > >>>>
> > >>>>  Let's say: everytime you add a switch and a if condition the code
> > >> splits
> > >>>>> in
> > >>>>> two and the fpga seems to execute both. It's all parallel. It's
> hard
> > to
> > >>>>> understand...
> > >>>>>
> > >>>>
> > >>>> Is the code public? I can attempt to explain it if you're
> interested?
> > >>>>
> > >>>> Regards,
> > >>>>
> > >>>> --
> > >>>> |              Mark McDougall                | "Electrical Engineers
> > do
> > >> it
> > >>>> |  <http://members.iinet.net.au/~msmcdoug>   |   with less
> > resistance!"
> > >>>>
> > >>>> --
> > >>>> Coco mailing list
> > >>>> Coco at maltedmedia.com
> > >>>> https://pairlist5.pair.net/mailman/listinfo/coco
> > >>>>
> > >>>
> > >>
> > >> --
> > >> Coco mailing list
> > >> Coco at maltedmedia.com
> > >> https://pairlist5.pair.net/mailman/listinfo/coco
> > >>
> > >
> >
> > --
> > Coco mailing list
> > Coco at maltedmedia.com
> > https://pairlist5.pair.net/mailman/listinfo/coco
> >
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco
>