SONY MEU-WX2 connect non SONY monitor with EDID adaptor? DIY EDID data spoof?

[solderfun]

The SONY MEU-WX2 is(was ) a professional video color processor and video standard converter/display device.

That was a lot. 

Basically you connect a video source (many inputs like SDI, composite, ...) and get DVI output. This device is usually used as a video display monitor on TV studios.

I want to connect a non SONY monitor to it.  :'(

The monitor needs to provide 1280x768. Which should not be a problem, unless there is no way to make set this resolution.

The device is quite picky and does not accept any monitor. I guess it detects proper display by the EDID standard. https://en.wikipedia.org/wiki/Extended_Display_Identification_Data

There are some adapter boxes like the https://www.extron.com/product/edid101d#features from a professional manufacturer like Extron available.

 Has anyone had some success with this approach?

Is there a DIY EDID data spoof box available?


Thanks a lot.

[tooki]

From reading the last two pages of this thread, I honestly think it’s probably not worth the effort to try:

https://shmups.system11.org/viewtopic.php?f=6&t=53953&start=60

It doesn’t seem like anyone has actually cracked the code as to what EDID it’s looking for.

[spostma]

You could make an EDID spoof box using these parts:
https://www.aliexpress.com/item/4000494867545.html
https://www.aliexpress.com/item/1005003086410568.html

[wraper]

If monitor does not support 1280x768 resolution, then EDID hack will not change that. You will get the signal from MEU-WX2 but monitor will display nothing or something like resolution not supported. Check monitor EDID first to see what it actually supports.
EDIT: I guess it's unlikely it supports such resolution in EDID. But I guess you could create custom resolution on PC, disable GPU scaling (otherwise it will just scale on GPU and output native resolution), and see if it actually works.

[tooki]

From the discussion I linked, it does not appear that copying the EDID data is successful. People have only managed to use other displays by actually connecting the matching Sony display in parallel.

[wraper]

The problem I've seen with dumping EDID on windows PC, it that only first EDID block is dumped, while there can be two or more. So it might be the issue there.

[wraper]

Also there may be an EEPROM on a non standard I2C address for EDID. Like in ATI DVI-HDMI adaptors which enabled HDMI audio over DVI output. They had an EEPROM on DDC lines but with address inputs set high so it does not interfere with EEPROM within a monitor.

[JacobPilsen]

Is it really that hard to eavesdrop on IC2 communication?

[tooki]

Is it really that hard to eavesdrop on IC2 communication?

Not at all. Reverse-engineering the data you record is the harder part.

Assuming the reports are correct, and off the shelf EDID spoofers don’t work here, it’s simply because Sony is performing communication on the I2C bus that goes beyond what the EDID specification calls for, so the spoofer only spoofs the standard part.

So you’d have to figure out which device is sending what, so that you can replicate the traffic.

[Tunertom]

Apologies for thread revival, just thought I'd add my findings incase anyone comes across this thread in future

I successfully unlocked the MEU units! - now can connect any display and output in any resolution

I have to say the original monitors are trash because of ancient CCFL backlights that make blacks grey at best - and these units paired with a decent monitor, IPS or OLED are amazing, I actually got rid of my OSSC and retrotink because this just trumps them all imo.

The difficulties in holding back unlocking the units come because the MEU's don't just check EDID, the monitors that are meant to be paired up with them operate on a parallel bus over the i2c lines, with multiple eeproms communicating across different address ranges

The solution? It requires sacrificing a Sony LMD monitor originally supplied with the MEU, I took the daughter board out of the LMD monitor, which contains all of the monitors logic and internalised this board within the MEU (there is plenty of room)

Said daughter board needs to be connected to various voltage rails to power up and then simply connect the SDA + SCL pins from the donor monitors daughter board directly to the SDA and SCL pins of the MEU motherboard, snipping the pins on the DVI connector so they aren't also connected to the target display.

Took me many hours/days of trial and error to come up with changing resolution output, but there is enough information in this post now for anyone technically minded to perform the same mod, if anyone would like me to elaborate further just message me

Hope everyone has a nice Christmas holiday!
Peace out!
Tom

[tooki]

I successfully unlocked the MEU units! - now can connect any display and output in any resolution

…

The difficulties in holding back unlocking the units come because the MEU's don't just check EDID, the monitors that are meant to be paired up with them operate on a parallel bus over the i2c lines, with multiple eeproms communicating across different address ranges

The solution? It requires sacrificing a Sony LMD monitor originally supplied with the MEU, I took the daughter board out of the LMD monitor, which contains all of the monitors logic and internalised this board within the MEU (there is plenty of room)

Said daughter board needs to be connected to various voltage rails to power up and then simply connect the SDA + SCL pins from the donor monitors daughter board directly to the SDA and SCL pins of the MEU motherboard, snipping the pins on the DVI connector so they aren't also connected to the target display.

Nice work. It’d be interesting to actually sniff the bus and figure out what the extra communication is doing, so that one can modify it without needing to scrap the original monitor.

I have to say the original monitors are trash because of ancient CCFL backlights that make blacks grey at best - and these units paired with a decent monitor, IPS or OLED are amazing, I actually got rid of my OSSC and retrotink because this just trumps them all imo.

A little niggle: what makes you think the light source is what determines LCD contrast?!? In LCDs, the light source is separate from the “light valve” element. The poor black point is simply because the LCD panel itself can’t block enough light. It’d be just as bad if it were lit by LEDs or any other light source.

The improvements in LCD contrast and black point happened coincidental to, but not due to, the advancements in white LED technology that matured enough to enable LED backlights to replace CCFL, eliminating the downsides of that technology (need for a high voltage supply, and aging characteristics). If LEDs hadn’t improved, we’d still have much higher-contrast LCDs now than we did back when CCFL ruled.

Now, of course if you’re talking about micro-dimmed LED backlights (i.e. backlights with relatively fine-grained local dimming), LED can do better. But in theory one could make arrays of small CCFL elements to do the same thing. And in a sense, we once did: that’s more or less what a plasma display is. (That’s an interesting thought experiment: what if we had combined plasma panels with LCD? Would that have given us further improvement in contrast and black point?)

[Tunertom]

Unfortunately sniffing the i2c bus is beyond my skill level, from what I can see though the monitors daughter board uses proprietary Sony chips and I know it communicates back and forth constantly, as disconnecting SDA or SCL lines after video output is live and working immediately results in black screen requiring reboot.

If someone more experienced wants to try sniffing the bus I'm more than happy to help.

Output resolution, refresh rate, serial numbers etc can be manipulated by flashing in standard EDID area of monitor but are not standard vesa values. Luckily not write protected so can be changed without hardware mod.

Regarding the ins and outs of the original backlight technology, you could be correct however I've yet to experience any older CCFL monitor that can actually show good black levels, with these particular displays the case could be that they are so terrible because they all have thousands of hours on them due to being used in a production environment and worn out, however the MEU units are built like Sherman tanks, I have 3 of them which I personally use now and couldn't be happier with them paired with modern displays.

[tooki]

Regarding the ins and outs of the original backlight technology, you could be correct however I've yet to experience any older CCFL monitor that can actually show good black levels, with these particular displays the case could be that they are so terrible because they all have thousands of hours on them due to being used in a production environment and worn out, however the MEU units are built like Sherman tanks, I have 3 of them which I personally use now and couldn't be happier with them paired with modern displays.

As I already explained, that is not because those had CCFL backlights, but because LCDs at the time simply weren’t as good.

Think of it like this: If someone said “cars released before CD players generally perform worse in crash tests than cars released with CD players, so CD players increase crashworthiness”, you’d rightfully dismiss that as nonsense. But the correlation is probably true, because cars that predate the CD player were from an era with significantly inferior crash requirements than cars released after.  But that is because crash requirements were tightened at the same time as CD players became popular. Concurrent technological development, not causality.