LKV373A HDMI streamer - my findings

[NiHaoMike]

The LKV373A is a really cheap way to capture HDMI at up to 1080p. At $33, the only cheaper capture devices I could find are $20 USB adapters with bad reviews.

Info page:
https://opentechlab.org.uk/videos:003:notes

Initial teardown and reverse engineering:
https://blog.danman.eu/new-version-of-lenkeng-hdmi-over-ip-extender-lkv373a/

Firmware collection:
https://drive.google.com/drive/folders/0B3mWuDyxrXyKZkxwYi1JNllENXc

I experimented with a few and here are my findings:
Use IPTV_TX_PKG_v4_0_0_0_20160427.PKG for the TX because it actually gives you a web UI to change settings. You will have to do a factory reset using the telnet trick (found on the reverse engineering page) after loading the firmware. The update UI is very badly designed since it will ask you to reboot the device as soon as the upload is finished, possibly bricking the device since it then takes a while to actually finish updating. Just leave it for a few minutes when it asks you to reboot it and it will reboot by itself.
For the encoder, there are two firmwares worth using:
Encoder_20160407_0942.bin gives full 1080p, but has more lag.
LKV373A_TX_V3.0c_d_20161116_bin.bin has less lag but doesn't give full 1080p.

I think in practice, the full 1080p firmware is going to be more useful.

The socat command given on the info page doesn't seem to work for me, but this does:

Code: [Select]

socat UDP4-RECV:5004,ip-add-membership=239.255.42.42:0.0.0.0 - | tee cap.ts | mplayer -

[cdev]

How do you like the picture quality and usability, generally?

Is it stable?

[NiHaoMike]

The quality is pretty good. The latency is too high for remote desktop use. Not sure how stable a stock unit will be since I upgraded the heatsinks in mine pretty early on.

[cdev]

watch out that it doesn't get too hot in there or your heatsinks may 'migrate" and eventually fall off, shorting something out.

The migration (not the shorting) happened to me with a video capture card I have.
I had never seen that before, the heatsink pad material softened up and the HS moved on several of them, luckily not far enough to fall off before I discovered it.

The computer's fan was set to slow itself down when the CPU was not working hard but the video capture card seemed to draw power and run hot all the time the computer was on. So i removed it.

[BradC]

Now to wait for them to come back into stock.

[amyk]

Next to no information on the Internet about the VCT6299/6289 ICs... despite the dates suggesting that they're almost 2 years old. My guess is that this is a 100% "shanzhai" device and whoever makes them doesn't want the IP/patent guys to find out who they are.

[NiHaoMike]

watch out that it doesn't get too hot in there or your heatsinks may 'migrate" and eventually fall off, shorting something out.

The migration (not the shorting) happened to me with a video capture card I have.
I had never seen that before, the heatsink pad material softened up and the HS moved on several of them, luckily not far enough to fall off before I discovered it.

The computer's fan was set to slow itself down when the CPU was not working hard but the video capture card seemed to draw power and run hot all the time the computer was on. So i removed it.

The upgraded "heatsink" is actually some an aluminum block and thermal pads to use the existing aluminum case as a heatsink. The block is glued to the case with JB Weld and isn't going anywhere. The thermal pads bridge the gap between the chips and the block. A big reason I did it that way is so I can still access all the pins on the chip in case I want to do more in depth hacking/reverse engineering later on.

[cdev]

That is a really great idea!


Did you take any photos as you did this? Thats what I was thinking I should consider with it.

watch out that it doesn't get too hot in there or your heatsinks may 'migrate" and eventually fall off, shorting something out.

The migration (not the shorting) happened to me with a video capture card I have.
I had never seen that before, the heatsink pad material softened up and the HS moved on several of them, luckily not far enough to fall off before I discovered it.

The computer's fan was set to slow itself down when the CPU was not working hard but the video capture card seemed to draw power and run hot all the time the computer was on. So i removed it.

The upgraded "heatsink" is actually some an aluminum block and thermal pads to use the existing aluminum case as a heatsink. The block is glued to the case with JB Weld and isn't going anywhere. The thermal pads bridge the gap between the chips and the block. A big reason I did it that way is so I can still access all the pins on the chip in case I want to do more in depth hacking/reverse engineering later on.

[NiHaoMike]

Not much to show, really. It was just a matter of trial and error to find the right size block that would fit, then add some thermal pads to bridge the remaining gap.