HDMI sources rejected in 3.2?

Hi All -

First of all, VERY happy to see this forum!

When using ATEM version 3.2 on a 1 M/E switcher, my Mac's HDMI input is not working as a source. I have downgraded to 3.1.2 until this has been resolved.

-Ed in Seattle

I've talked with BMD and they said that they have had several calls about this issues. What they said is that the switcher is looking for a 59.94fps signal but if the signal is 60 then that's fine too. Computers use a normal rate of 60Hz for monitors. 60Hz = 60fps. But with this new update that tolerance is tighter to where the 60Hz does not match. They are working on trying to fix it.

Yes, here is the same situation. I can't use my HDMI from Kramer VP-434 (60Hz for HDTV resolutions) to ATEM 1 M/E. I had downgraded to 3.1.2 for work.

Johnny in Taiwan

I've done extensive testing on the 3.2 update on my 1ME and it's impact on HDMI devices. I posted my results on the Atemusers.com site.

Whatever was tightened up needs to be un tightened up as many of us now have various scalers and converters and Apple TVs that no longer function with the 1ME.

I do hope the reports we have heard that this will be resolved in a soon to be released firmware update is true as I now have a lot of expensive devices that are now useless. These are devices that previously worked and were an essential part of my gear used in live events.

The fact is that all these devices still function when connected to my TVS (with 3.2 firmware). Hopefully this points to an easy fix.

Black Magic Designs...Thanks for listening to your loyal customers.

[Darren Lew]

Hi everybody and welcome to the forum.

I've noticed that some of the ATEM users are having a bit of trouble using 60Hz as an input.

So what has happened is this. We did the 3.2 update, which gave everyone a software audio mixer, and tweaked our inputs. We had no idea that people were using 60Hz as an input, What a way to find out! So now that we know, we are going to put this right back in there.

I hope that we can get you guys back up and running asap. In the mean time if you can tweak your video card output to 59.94, our supported format, then that will do as a work around.

Cheers

Darren.

Interestingly, 3.2 resolved compatibility issues with the Hyperdeck Studio running in PAL via HDMI.
Or at least that is my experience on the 1ME

Tim

Hi Darren

There are also some issues with the EDID that the ATEM provides - it is not supplying all of the possible resolutions to computers correctly. Specifically the SD resolutions are incorrect and (depending on graphics card) it can be impossible to force the correct resolution to output (getting graphics cards to output interlaced SD is troublesome - in many ways it would be better if the ATEM could handle 720x576 @ 50p - the miniconverters seem to cope ok with this resolution)

Also it would be really much nicer if the EDID could be dynamic so the "preferred resolution" in the EDID was updated to match the current ATEM resolution. It's annoying that each time you connect a PC it defaults to 1080i 60 (err 59.94! ) when we are in PAL land and simply never use 60hz for anything.

I sent these comments to Jess the other day. Is she still the best contact for ATEM development? I've not heard much from her recently and it's a little concerning that ATEM development seems to of become side tracked away from the functionality which was originally announced. Whilst I'm sure some people might like the audio mixer it is a serious concern that the 2ME still doesn't have basic functions, which are advertised as part of the switcher on your website.

Cheers
Tom

Also it would be really much nicer if the EDID could be dynamic so the "preferred resolution" in the EDID was updated to match the current ATEM resolution. It's annoying that each time you connect a PC it defaults to 1080i 60 (err 59.94! ) when we are in PAL land and simply never use 60hz for anything.

Dynamic EDID would be great, or at least EDID support for PAL.

I have had no problem with Nvidia cards but could not get an ATI laptop to output 576i despite spending hours messing with the EDID data.
The only way this would work was using Ubuntu without the ATI driver installed and adding a modeline.

Tim

I am really happy to see this forum, great job

I have seen a few issues with my TVS that based on the above posts might be that the Notebook I was using had an ATI 9800m card. My Notebooks with Nvidia 9800gt and gtx cards work great. On the ATI cards, I have the following issues:
Using a DVI to HDMI cable in need to restart the Notebook in order for the signal to display (this might be related to the EDID issue???)

If I use a BMD DVI extender on the ATI 9800m equipped notebook, I cannot under any circumstances get the SDI signal to display even after simply replacing the DVI-HDMI cable with the DVI extender.

On my Nvidia 9800gt/gtx, none of these problems occur, the notebook is recognized on the HDMI input, via the DVI-HDMI cable, and also Via the DVI extender using the SDI cable (the DVI extender in all cases is set for video instead of HDlink)

These are all using 1080i

Again Great to have this forum
Sharyn

I hope that we can get you guys back up and running asap. In the mean time if you can tweak your video card output to 59.94, our supported format, then that will do as a work around.

Darren. Thanks for confirming the issue and that you'll be puting 60Hz back in there. Much appreciated.

Just a note on the video card output. The video card compatibility was (for me) the least problematic - however my card (and many others) do not seem to have a 59.94 option. I got mine working with 3.2 by setting it to 59hz (but is that really 59.94 or is it 59 and the 3.2 firmware still accepted it?)

My biggest issue is with other scalers/converters where they do not appear to have any 59 or 59.94 option listed.

And... then there is the fact the TVS and 3.2 was not impacted by the 60Hz issue.

In any case - I will be very happy to see the 60Hz support being formally recognized and supported. Thank you!

I just decided to try a Atlona AT-HD560 HDMI this is supposedly a scaler/converter. I have a HP notebook with HDMI out 1600x1200, and I thought the unit would scale it to 1920x1080i at 59. what I have found is that it does do the 59i conversion, but the signal seems to remain at 1600x1200 OR if it pads the signal up for a full 1920x1080i the TVS still will not see it. I was hoping that this scaler/converter might be a solution for some interfaces. Interestingly when I connect ther Atlona HD560 directly to the Vizio TV it works fine. SO possibly it has the same problem re edid as I have outlined below???

SO I then decided to do more tests. I decided to connect my notebook (HP g71-441nr integrated video directly to my Hdmi tv. Interestingly when directly connected, the 1920x1080 resolution option apprears when directly connected to the TV via HDMI, but when I replace it with the ATEM TVS and BMD device shows, the resolution options are NOT available.

SO I am not the engineer, and probably not the most experienced person by a long shot, BUT it would seem to me that POSSIBLY, the ATEM TVS is not returning a "proper" edid so for some reason the information the device attached needs to receive to make the proper resolution decisions is MISSING??

Is this possible? Since it seems to work with a HDMI TV (VIZIO) and all the proper resolutions appear

I would be interested to see what more experienced people and BMD thinks???

THANKS
Sharyn

I posted more regarding this on the ATEMUSER Forum, but again people more experienced than I might find this stuff possibly usefull


I have a tvs and a series of converters/ and notebooks

What I am seeing is that I have no problem with my Nvidia 9800gt or gtx cards at 1080i 60, the connect and display immediately directly connected to the TVS via DVI to HDMI cable OR DVI extender via SDI
]
On my ATI 9800m cards, the only way, to get the DVI to HDMI cable connection to work is to restart the system. Since this does not work with the DVI extender via SDI there is not ability In my case to get the DVI via SDI extender to work

I tried my TVone 622 (analog to HDMI) and it works fine with the TVS in 1080i60 and 720p60 mode (after I remembered to change the setting in the control software)

I then tried to use a HP notebook with the integrated Intel chip set and hdmi out. If I connect it directly to the TVS, the BMD device is detected but the 1920x1080 resolution is not available, so it does not work. I then had a ATLONA 560 scaler for hdmi available and tried that thinking that it would take the normal 1600x1200 (the resolution of the internal monitor on the Notebook) and convert it, this did not work

I then decided to test connecting the HP directly to the VIZIO monitor, and with this connection, the VIZIO shows up, and the 1920x1080 resolution also shows up as an option and works.

SO I was wondering if part of the problem might be that the TVS IS NOT DOING THE EDID PROPERLY. I am guessing that if for some reason the correct 1920x1080 resolution is not being given back to the connecting device (notebook converter etc) that that MIGHT be a cause of some problems? I don't have any sort of EDID doctor or what ever but since these devices work properly when directly connected to a monitor, and have the correct resolution options (granted they might be 59.94 vs 60) if some of the root cause is a EDID problem?

Someone more experienced might be able to comment on this.
On this forum there is an acknowledgement of a problem but it seems to be more associated with just allowing 60 vs 59.94 which is certainly an issue for some, but I wonder if at the same time there might be an EDID issue interestingly when I decided to WIKI EDID it APPEARS that there is NO provision for 60 or 59.94 independent of one another, the two options seem to be shown for each resolution. see the Bottom section of
http://en.wikipedia.org/wiki/Extended_d ... ation_data
br />

Microsoft shows a way to force a edid in widows7 that could override what the ATEM is providing. For the brave:-)
http://msdn.microsoft.com/en-us/windows ... 87330.aspx

which is then related to the issue that WIN7 does not record the full EDID in the registry (only 128bytes instead of the full 256, which MAY explain where the graphics card driver might be doing a better job of getting and storing this information and explain why NVIDIA cards seem to have less problems that ATI and Intel?????? Nvidia might be doing a better job of this??????

http://social.technet.microsoft.com/For ... ead/9294...

thanks
Sharyn

[Darren Lew]

"Tom_Bassford"

Hi Darren

There are also some issues with the EDID that the ATEM provides - it is not supplying all of the possible resolutions to computers correctly. Specifically the SD resolutions are incorrect and (depending on graphics card) it can be impossible to force the correct resolution to output (getting graphics cards to output interlaced SD is troublesome - in many ways it would be better if the ATEM could handle 720x576 @ 50p - the miniconverters seem to cope ok with this resolution)

Also it would be really much nicer if the EDID could be dynamic so the "preferred resolution" in the EDID was updated to match the current ATEM resolution. It's annoying that each time you connect a PC it defaults to 1080i 60 (err 59.94! ) when we are in PAL land and simply never use 60hz for anything.

I sent these comments to Jess the other day. Is she still the best contact for ATEM development? I've not heard much from her recently and it's a little concerning that ATEM development seems to of become side tracked away from the functionality which was originally announced. Whilst I'm sure some people might like the audio mixer it is a serious concern that the 2ME still doesn't have basic functions, which are advertised as part of the switcher on your website.

Cheers
Tom

Hi Tom,

Thanks for taking the time to post on our new forum.

We appreciate how patient most users have been in waiting for some features to come along and I am happy to say that in the very near future one of our most anticipated and requested features will be with us. Even while some exciting features are being released others are being developed in parallel. So new features will keep on coming as they are ready.

Cheers

Darren.

Hi Darren.

One thing that might be worth considering is the ability via the console software or another utility to have some diagnostic information, ie look at the EDID or at least the information it is providing, if there any information stored in the Atem as to why certain inputs are being rejected etc and any other functions and settings in the ATEM that might be useful to debug problems

Sharyn

Hi Darren, do you have an estimated date in which this bug be fixed, not only our mac´s stop being compatible but our PC´s, scalers and some media players too, this is a big problem because we have contracts with our clients that rely in our set up and we cant accomplish what we used to, this equals to loosing clients and getting also a bad reputation.

I appreciate your response thank you.

3.3 was released today, but I don't see this listed in the release notes as being fixed.

-Ed

[Joshua Helling]

3.3 was released today, but I don't see this listed in the release notes as being fixed.

-Ed

This update only included Supersource to the 2ME and some enhancements to the SDK.

Hi Joshua -

Thanks for the speedy update. While SuperSource sounds like a great feature, I need to retain my computer inputs. Looking forward to this issue being resolved.

-Ed

[Joshua Helling]

Yup yup, totally understand.

In the mean time you can see if your computer will do 59Hz output (some say they do, but really just do 60Hz anyways) or you can roll back to 3.1 until we can get that feature back in (which I assure you, we are working on).

Has the 60HZ issue been resolved in the 3.5 firmware?

Has the 60HZ issue been resolved in the 3.5 firmware?

Yes. It was fixed in 3.4.

Cool, thanks!