Just to clarify, this is the absolute cheapest encoder that Resi offers. If you buy the server grade encoders they are bulletproof. I've done the exact same test where I pulled the ethernet cable out of the wall. It kept streaming perfectly. I have a 2 minute delay set on purpose for something like this, and I legit pulled the plug out for about 1:45, plugged it back in and it didn't miss a beat.
@@johnmarshalward, I have been using 2 Atem Mini Pro on the same network used at the exact same time to broadcast our live stream service for the last 3 years. I honestly haven't had any issues with uploading.
We were having terrible problems with our ATEM until we upgraded the firmware and did a reconfiguration of the streaming settings (which got significantly changed in the updated firmware). We did tons of testing and until we made this change we were having streaming issues every time. Sundays were bad when the building was full. But practice night when no one was on the network were also bad. Once we did the update we had zero issues. The cache sits near zero (as indicated by the fact that when you push the off button it stops immediately) Note: on the ATEM the cache doesn't show a problem until it reaches 40%. So if you are using an ATEM Mini, upgrade the firmware and reconfigure your stream settings.
Interesting video! Couple of things: I understand that theres no coincidence, but it still feels a little weird thst the device "winning" the competition is rhe sponsor of the video. I would've found it interesting to see how srt (e.g. out of vmix) as streaming protocol compared to rtmp (pro presenter).
Good video. There are a few things to discuss and consider which color your test. You have to take into account. Internet speed, network hardware and configuration, cable type (Ethernet or fiber), distance between the encoder and the connection point to the router , other traffic on the network. Your video resolution and settings are a factor. There are also a number of additional factors and all of this should be addressed and planned for to mitigate issues. Furthermore. How the video is distributed and the solution for this is also something to consider.
Great video! I really wish you would have included the black magic web presenter HD with this comparison. This is what we are using now. Originally we were streaming with the Atem. Our upload speeds were not good and crashed frequently so in an attempt to remedy the poor internet service we upgraded to the web presenter because the video could be scaled down and as a result the bit rate would be reduced by half (about 1.6 up). Even with the bit rate so low we still see crashes every 2-3 weeks. I just placed all the blame on the connection but this makes me wonder if another encoder would be more stable and able to handle the unreliable internet connection better.
A little surprised by the Resi results. Particularly on the disconnect and packet loss tests. The whole point of their system is to cache and fill in gaps bedore streaming out. I'm curious if there is a difference between the mini unit and their other encoders?
I would be curious if he is using resi in rtmp mode or rsp protocols. I believe you can still add a delay time to resi when it is rtmp mode so that is not indicative of the mode he is using when he mentions the test parameters compared to the boxcast. We use a e1200 for our encoder from resi and use propresenter on an old system with a decklink card as a back up. Everything is hardlined on gigabyte switches I have yet to have a failure like this ever. The only failures I have had has been from either user error or actual full blown hardware issue. This seems more like a commercial for h2.65 more than anything else. The dynamic bit rate adjustment is cool. I love Jake and all he does for the church tech community. I think this is as good a test as he could do with what he has but I really think either the boxcast or the resi platform are great depending on church budget. If a church is able to go to a full-blown server grade with resi I don't see them having these issues but they are very different products focused on different customer needs. Not saying more expensive is better I am just saying they are marketed to different needs. Bot companies have always had great support. Again, both are good just different.
It would have been interesting to see how a Liveu Solo would have performed in this test considering one it's main features is being able to distribute the load over muiltple networks.
The Spark encoder doesn't require a dedicated computer to stream, since it's a standalone encoder. We like to use it for events where reliability is critical, if we have any minor outages, Spark will keep streaming. This is especially helpful for longer events. It also has HEVC compression, which means you can stream with twice the quality using less bandwidth. I also just really like that when I'm producing an event it's easy to keep tabs on the streams status just by glancing over at the Spark, since the LED's and LCD show quick status on how the stream is doing. I'm also just a big fan of using dedicated hardware, since I've run into too many issues relying on a computer in the past. OBS is great though, it's hard because it's not really and apples to apples comparison, and OBS and Spark are likely to be used with very different streaming setups.
@@joshyclemI had to drop my bitrate down from 2000 Kbps to 1900 Kbps and the FPS value from 30 down to 24 NTSC in order for OBS to not automatically stopping the livestream. I had been running at bitrate 2000 and 30 FPS for several months until about a month ago OBS would just stop streaming.
@@MarcusGaines-mh4ol Oh that's tough! It might be worth looking into Spark, we've streamed at 1.5mpbs on cell networks and still gotten great picture quality.
I'm curious to know if variable bitrate encoding was used on some encoders vs constant bitrate encoding. If you have VBR enabled, the bitrate you set is really just a target and it's going to vary a decent amount based upon the amount of video data that the encoder has to crunch. The impact on different encoders is going to vary wildly too... Propresenter is already fighting for CPU resources, so it wouldn't be able to handle CBR encoding if it needs to process a lot of changing video at once. There are a lot of variables in video streaming.
A computer with OBS wins every time! Everyone's answer always seems to be, "throw money at it with a new device and subscription"! When in all honesty you need to reach out to your tech people. They should be in the loop. God sent them your way to help you. Allow them to flex their gift! Too many times I have seen where something is implemented without understanding and no AV, tech person or anyone who even likes tech was ever in the discussion. It was just a question to a service provider and all they said was, "yes we do that.". No options for quality, performance, endurance. Sorry I am going to get off this box now. Continue sir!
Is anybody else having problems with their Spark encoder's local recordings? Since day one, they have been unusable for us. The framerate is awful, and the playback looks like it's in slow motion. We're outputting 30fps from our switcher, and both the stream and local recording we make to our Hyperdeck Mini look fine. Support has been unable to provide any answers. Very frustrating.
$1000 for the boxcast plus subscription seems incredibly steep. I could build a computer with a intel b570, stream in av1 instead of h.265, and have no subscription fee for that price with better quality of video and less bandwidth needed. Even just get a EGPU enclosure and slap a b570 in it.
My church has been looking to move away from Pro7 & Restream. Av1 seems like the most ideal option for encoding especially since our upload speeds are quite low. Lower bandwidth and better compression. I’ve been considering a dedicated PC build for streaming and producing using vmix.
Just to clarify, this is the absolute cheapest encoder that Resi offers. If you buy the server grade encoders they are bulletproof. I've done the exact same test where I pulled the ethernet cable out of the wall. It kept streaming perfectly. I have a 2 minute delay set on purpose for something like this, and I legit pulled the plug out for about 1:45, plugged it back in and it didn't miss a beat.
Oddly, the Atem isn't included in the test. It is literally on the left of the desk.
Can you say "Cache Full" ?
We can do that next time. But I imaging it would be same result as propresenter.
The ATEM is exactly where it needs to be. Which is anywhere it wants as long as it's not being used to upload a livestream.
@@johnmarshalward, I have been using 2 Atem Mini Pro on the same network used at the exact same time to broadcast our live stream service for the last 3 years. I honestly haven't had any issues with uploading.
We were having terrible problems with our ATEM until we upgraded the firmware and did a reconfiguration of the streaming settings (which got significantly changed in the updated firmware). We did tons of testing and until we made this change we were having streaming issues every time. Sundays were bad when the building was full. But practice night when no one was on the network were also bad. Once we did the update we had zero issues. The cache sits near zero (as indicated by the fact that when you push the off button it stops immediately) Note: on the ATEM the cache doesn't show a problem until it reaches 40%.
So if you are using an ATEM Mini, upgrade the firmware and reconfigure your stream settings.
Interesting video! Couple of things: I understand that theres no coincidence, but it still feels a little weird thst the device "winning" the competition is rhe sponsor of the video.
I would've found it interesting to see how srt (e.g. out of vmix) as streaming protocol compared to rtmp (pro presenter).
Good video. There are a few things to discuss and consider which color your test. You have to take into account. Internet speed, network hardware and configuration, cable type (Ethernet or fiber), distance between the encoder and the connection point to the router , other traffic on the network. Your video resolution and settings are a factor. There are also a number of additional factors and all of this should be addressed and planned for to mitigate issues. Furthermore. How the video is distributed and the solution for this is also something to consider.
Great video! I really wish you would have included the black magic web presenter HD with this comparison. This is what we are using now. Originally we were streaming with the Atem. Our upload speeds were not good and crashed frequently so in an attempt to remedy the poor internet service we upgraded to the web presenter because the video could be scaled down and as a result the bit rate would be reduced by half (about 1.6 up). Even with the bit rate so low we still see crashes every 2-3 weeks. I just placed all the blame on the connection but this makes me wonder if another encoder would be more stable and able to handle the unreliable internet connection better.
A little surprised by the Resi results. Particularly on the disconnect and packet loss tests. The whole point of their system is to cache and fill in gaps bedore streaming out. I'm curious if there is a difference between the mini unit and their other encoders?
I have streamed out using resi over cellular connections and had smooth results. This was using their modified Nuc encoders and a cradlepoint.
I would be curious if he is using resi in rtmp mode or rsp protocols. I believe you can still add a delay time to resi when it is rtmp mode so that is not indicative of the mode he is using when he mentions the test parameters compared to the boxcast. We use a e1200 for our encoder from resi and use propresenter on an old system with a decklink card as a back up. Everything is hardlined on gigabyte switches I have yet to have a failure like this ever. The only failures I have had has been from either user error or actual full blown hardware issue.
This seems more like a commercial for h2.65 more than anything else. The dynamic bit rate adjustment is cool.
I love Jake and all he does for the church tech community. I think this is as good a test as he could do with what he has but I really think either the boxcast or the resi platform are great depending on church budget. If a church is able to go to a full-blown server grade with resi I don't see them having these issues but they are very different products focused on different customer needs. Not saying more expensive is better I am just saying they are marketed to different needs. Bot companies have always had great support. Again, both are good just different.
I really miss the ATEM and OBS. I am actually about to update our church’s broadcast system, so this would be very interesting.
It would have been interesting to see how a Liveu Solo would have performed in this test considering one it's main features is being able to distribute the load over muiltple networks.
How does OBS compare with the boxcast spark?
The Spark encoder doesn't require a dedicated computer to stream, since it's a standalone encoder. We like to use it for events where reliability is critical, if we have any minor outages, Spark will keep streaming. This is especially helpful for longer events. It also has HEVC compression, which means you can stream with twice the quality using less bandwidth. I also just really like that when I'm producing an event it's easy to keep tabs on the streams status just by glancing over at the Spark, since the LED's and LCD show quick status on how the stream is doing. I'm also just a big fan of using dedicated hardware, since I've run into too many issues relying on a computer in the past. OBS is great though, it's hard because it's not really and apples to apples comparison, and OBS and Spark are likely to be used with very different streaming setups.
@ Thank you for your reply and explanation, it gives me something to think about especially since I have been having trouble with my livestream.
@@MarcusGaines-mh4ol Sure thing! What kind of issues are you running into?
@@joshyclemI had to drop my bitrate down from 2000 Kbps to 1900 Kbps and the FPS value from 30 down to 24 NTSC in order for OBS to not automatically stopping the livestream.
I had been running at bitrate 2000 and 30 FPS for several months until about a month ago OBS would just stop streaming.
@@MarcusGaines-mh4ol Oh that's tough! It might be worth looking into Spark, we've streamed at 1.5mpbs on cell networks and still gotten great picture quality.
Would love to hear responses from Resi and Boxcast devs on this.
i would like to see results of the AJA helo, Epiphan pearl, matrox monarch, and a Black Magic web presenter
Why didn't you include the ATEM that was sitting right there?
I'm curious to know if variable bitrate encoding was used on some encoders vs constant bitrate encoding. If you have VBR enabled, the bitrate you set is really just a target and it's going to vary a decent amount based upon the amount of video data that the encoder has to crunch. The impact on different encoders is going to vary wildly too... Propresenter is already fighting for CPU resources, so it wouldn't be able to handle CBR encoding if it needs to process a lot of changing video at once. There are a lot of variables in video streaming.
what tool did you use to simulate the packet loss?
what about the yolobox series ?
We use Vmix and Amplify for broadcast and don't have ANY problems.
A computer with OBS wins every time! Everyone's answer always seems to be, "throw money at it with a new device and subscription"! When in all honesty you need to reach out to your tech people. They should be in the loop. God sent them your way to help you. Allow them to flex their gift!
Too many times I have seen where something is implemented without understanding and no AV, tech person or anyone who even likes tech was ever in the discussion. It was just a question to a service provider and all they said was, "yes we do that.". No options for quality, performance, endurance. Sorry I am going to get off this box now. Continue sir!
I am still waiting for 4K encoder with HDR BT2020, support and AV1 support.
Is anybody else having problems with their Spark encoder's local recordings? Since day one, they have been unusable for us. The framerate is awful, and the playback looks like it's in slow motion. We're outputting 30fps from our switcher, and both the stream and local recording we make to our Hyperdeck Mini look fine. Support has been unable to provide any answers. Very frustrating.
Hi there! We're really sorry to hear that you're running into this issue, we'll be reaching out shortly to you to try and get this resolved.
$1000 for the boxcast plus subscription seems incredibly steep. I could build a computer with a intel b570, stream in av1 instead of h.265, and have no subscription fee for that price with better quality of video and less bandwidth needed. Even just get a EGPU enclosure and slap a b570 in it.
My church has been looking to move away from Pro7 & Restream. Av1 seems like the most ideal option for encoding especially since our upload speeds are quite low. Lower bandwidth and better compression. I’ve been considering a dedicated PC build for streaming and producing using vmix.
@@haydenvarner2501 would be way cheaper than a locked down box like what Jake is promoting here
Where are you streaming to that accepts AV1? Twitch?
@ RUclips and vimeo both accept AV1
We use the basic BoxCast encoder - it's only $399 and works great for our environment.
Box cast won
what about zoom