Tuesday, March 19, 2013

Fixed the DVI / pin-16 hotplug dilemma

As with a lot of mod'ing or (dare I say it!) bodging solutions you need to find a nice connector or pre-made cable to base your fix on. If you look back at the problem we've been having with Media Composers switched across different Amulet heads then you'll recall it wasn't an EDID issue, rather one of Windows detecting a monitor change; Amulet does the right thing, it's Avid that's the problem.

The fix is easy; you need to tie pin-16 (hot plug detect) to Vcc (+5v on pin 14) via a 1K resistor;




The best mod'able pre-made cable that is suitable is one of these from Lindy.  Now I just need to knock up a dozen for the customer!

15 comments:

Wim Brain said...

I also have a problem with screen detection on dvi, my view isn't going through a extron matrix. It only goes through when i reset my screen. Do you thing your fix will work?

Phil Crawley said...

Well it'll either be DVI detection - in which case the Lindy DVI-spoofers will do the job. If it's a hot-plug detection issue then pin-16 is the solution. It'll cost you less than £50 to buy the bits and try both.

Wim Brain said...

I tried the 1K solution and it works perfect. I haven't place him in the connector, because its a complete closed one, but i did it on the cable, now i have to fix the shield :( But thanks for the solution.

Phil Crawley said...

Brilliant
http://www.lindy.co.uk/05m-premium-gold-dvi-d-dual-link-cable/37037.html
Is the part we used; we had to get ten Avids working and possibly another fifty! In any quantity those cables are not the listed price - around £12 each and they are easy to re-work.

praternatural said...

Hi Phil, I am an EE student working in Atlanta, Georgia and I'm a fan of your podcast.
I've encountered the dreaded DVI/HDMI hotplug detect problem in my work and got wind of your DVI Spoofer featured on the EEV Blog Mailbag episode. Do you have any more information on your DVI Spoofer such as how much you're selling them for and where I can get some?

Phil Crawley said...

Well it turns out the company who got them made for me in their "one-hung-low" factory in China are selling them as well!

http://www.lindy.co.uk/cables-adapters-c1/audio-video-c107/dvi-c110/dvi-d-male-to-dvi-i-female-adapter-with-1k-resisitor-for-hot-plugging-p7914

If that URL is mangled the part number from Lindy UK is 41082

Phil Crawley said...
This comment has been removed by the author.
Ryo said...

I know this is old thread and I got here from the Intel support forums which it is still going on about this problem today!

If you're still monitoring this, I was wondering if I can install the resistor inside the monitor instead of the cable? I would assume technically it is the same thing but you may have thought of something that I have no knowledge of if I did it this way.

Any info would be greatly appreciated.

Ryo said...

Hi,

I made a comment earlier about doing this on the monitor side. Well, I went ahead and did it anyway because I was sick and tired of my display going out.

Just let you know it works perfectly so far. I used a 1.2k resistor instead because I didn't have 1k on hand. I figured more couldn't hurt.

In any case, just wanted to say thanks.

Phil Crawley said...

From my point of view modifying the monitor is a danger; but having a back-back adaptor or a short cable is the answer. If it's your own display then fair enough but I have to provide customers with a solution that won't confuse them!

Peter Mauro said...

Hi Phil,
I tried your fix , also using a 1.2 K resistor , it did not work.
I have to reload the Intel Graphics 4000 driver every time I start up from power down , sleep or hibernate.
Then my ASUS monitor is detected, and the DVI input works fine.
Have you found any other fix?

Peter

Phil Crawley said...

Well in the end I got sick of making cables for this (a lot of my customers need this to make a particular pice of software behave itself) and so I had a Chinese factory make me 500 DVI male-female adaptors that tie pin-16. BUT, remeber - all this stops is the graphucs card being able to detect a change of monitor, it does nothing to the EDID profile. For that I've found these to be useful;

http://www.lindy.co.uk/audio-video-c2/matrix-switches-splitters-c166/edid-ddc-adapter-for-dvi-displays-p6699

hks1966 said...

I posted this fix on another thread of yours a while back and don't know if you got it, so here it is again just in case as I am re-visiting the issue on some other Avids today and had to remind myself what to do. You can load the EDID in from file which will make MC ignore a monitor dis-connect/re-connect. Works for me. Only issue I can see if a lower scanning monitor is plugged in and can't display the file preset scan rate. It would be a case of delete the file I guess, and reboot.

https://sites.google.com/site/ebobster/stuff/displayportblanking

Phil Crawley said...

I wouldn't recommend a software fix; especially with Avid - it'll break on the next revision of Media Composer for sure.
I've now supplied more than a hundred of these hardware adaptors to folks with Avid and it really is the most solid way of stopping MC detecting monitor changes.

Johan Elliot said...

Thank you Phil!
Your 1k resistor fix is brilliant!
I soldered mine into the monitor, now it all works like magic.

You would think that Intel would care enough to make their drivers backward
compatible with older monitors that does not have the "Hot Plug"- signal?
It is not that they don't know how, their latest driver has a control panel
with a "Detect" button that does initialize such a monitor.
At PC startup however, it still does not work?!

Thanks again, Mate :-)