E1.31 Questions [Beta Released 6/8/12]

[Guest wbottomley]

Bingo Fasteddy. I wonder how many disappointed people they'll be when the season rolls around? As in their PC is not fast enough?

[edvas69]

wbottomley wrote:

Bingo Fasteddy. I wonder how many disappointed people they'll be when the season rolls around? As in their PC is not fast enough?

This why im raising this issue now so people with big plans may at least have in the back of their minds that they may have to also factor in a computer upgrade to help improve their sequencing experience.


I was one of those last year that upgraded to an i7 3.4ghz system with 16gb of ram so i could sequence without major frustration (be it with another package) as my dual core 3ghz system didnt cut it anymore.

I think nearly all the guys that did 5K plus sequences last year have upgraded their systems to cope with the added load of sequencing with a grid based sequencer. More data, improved GUI and effects all take its toll and this is where people should now be creating their large sequences so they can see if their sequencing experience is not slowed by waiting for the processor to crunch the numbers.



LOR S3 may perform better than other packages, im not sure as I dont have S3 (only S2) so again test, test and test some more to make sure you can do what you want with the computer hardware you have.

[Charles Belcher]

Note to all:

Our 2010 installation at the State Fair of Texas used 22 universes of DMX and we custom built an eight core processor desk top to handle the loads.

This brief YouTube video shows 15 CCR's spaced on 16' centers. The entire installation in 2010 was 136 CCR's and is being expanded this summer to include a total of 448 CCR's. In fact, we are having to duplicate universes in areas which are not in the same line of sight in order to keep processing down.

Charles

[Charles Belcher]

John,

You said in your first post...

" You can test using any hardware that supports the standard."

Does this include wireless transmission?

Charles

[edvas69]

Charles Belcher wrote:

John,

You said in your first post...

" You can test using any hardware that supports the standard."

Does this include wireless transmission?

Charles

Wireless transmission of E1.31 is not recommended mainly due to the fact that E1.31 has very limited error correction and this may allow packets of data to arrive out of order or not at all which may have adverse effects on the display.

You can still use E1.31 over a wireless network, but just be aware that you may get some odd effects depending on how good of a wireless connection you have

[Denny]

edvas69 wrote:

wbottomley wrote:

Bingo Fasteddy. I wonder how many disappointed people they'll be when the season rolls around? As in their PC is not fast enough?

This why im raising this issue now so people with big plans may at least have in the back of their minds that they may have to also factor in a computer upgrade to help improve their sequencing experience.


I was one of those last year that upgraded to an i7 3.4ghz system with 16gb of ram so i could sequence without major frustration (be it with another package) as my dual core 3ghz system didnt cut it anymore.

I think nearly all the guys that did 5K plus sequences last year have upgraded their systems to cope with the added load of sequencing with a grid based sequencer. More data, improved GUI and effects all take its toll and this is where people should now be creating their large sequences so they can see if their sequencing experience is not slowed by waiting for the processor to crunch the numbers.



LOR S3 may perform better than other packages, im not sure as I dont have S3 (only S2) so again test, test and test some more to make sure you can do what you want with the computer hardware you have.

Is there anyway to come up with a rough suggestion on CPU speed/number of cores/RAM, etc? Perhaps someone more computer literate than I could make minimum/starting suggestions based on something like 5 universes, 10 universes, 25 universes, etc.

[edvas69]

Denny wrote:

edvas69 wrote:

wbottomley wrote:

Bingo Fasteddy. I wonder how many disappointed people they'll be when the season rolls around? As in their PC is not fast enough?

This why im raising this issue now so people with big plans may at least have in the back of their minds that they may have to also factor in a computer upgrade to help improve their sequencing experience.


I was one of those last year that upgraded to an i7 3.4ghz system with 16gb of ram so i could sequence without major frustration (be it with another package) as my dual core 3ghz system didnt cut it anymore.

I think nearly all the guys that did 5K plus sequences last year have upgraded their systems to cope with the added load of sequencing with a grid based sequencer. More data, improved GUI and effects all take its toll and this is where people should now be creating their large sequences so they can see if their sequencing experience is not slowed by waiting for the processor to crunch the numbers.



LOR S3 may perform better than other packages, im not sure as I dont have S3 (only S2) so again test, test and test some more to make sure you can do what you want with the computer hardware you have.

Is there anyway to come up with a rough suggestion on CPU speed/number of cores/RAM, etc? Perhaps someone more computer literate than I could make minimum/starting suggestions based on something like 5 universes, 10 universes, 25 universes, etc.

This is very subjective and depends on many factors, the length of the sequence, the timing intervals, the % of different types of effects used and a few more other factors all go to make up how your computer will perform, so it is really hard to give any firm measurements in performance and thats why i say test, test and test some more.

[Guest wbottomley]

Fasteddy... you couldn't have timed these posts any better. When the last quarter arrives... I bet all kinds of issues will arrive.

Great information!

[edvas69]

wbottomley wrote:

Fasteddy... you couldn't have timed these posts any better. When the last quarter arrives... I bet all kinds of issues will arrive.

Great information!

This is when many of those who can help become to busy to spend a great deal of time helping. I know the 3 months leading up to showtime i end up putting much less time into the community as i need to focus on getting my own display done and this would be the case for many of the guys that have been around that have any experience.

[aussiephil]

With so much conversation happening on various forums it's good to see some real information being released.

999 Universes, a nice marketing ploy but in a grid based sequencer it will never be usable regardless of processor memory etc.

Now there was a question about hardware required. This can only be answered in relation to LOR once S3 with E131 support is released and someone pushes the outer limits and determines the real requirements. The actual processing of the data even at 25mS timings is a low overhead activity for just about any modern processor.

Going to have a lot of Pixels and you don't have a modern i5, i7, AMD 8Core then plan on doing a computer upgrade as all the indications with other software suggest that this is a requirement to ensure good performance.

I personally have run 8k to 10k channels the last two years (not LOR) using E131, 2011 show had 26 universes used. 2012 will see the channel count jump to 20,000 plus and another change in software direction to ensure performance is good.

Different software packages have over the last couple years shown differing hardware requirements and this is why we will need someone to real world test the upper requirements for LOR S3.

DON'T leave it to November/December to test and remember that just because you could sequence a single track that you can play back a complete show....... TEST the show playback early.

OH on BIG thing... with E131 you don't need the hardware plugged on the far end to test the performance envelope. This assumes true E131 compatibility.

I am also pleased to see that LOR S3 people will be able to use standardised Pixel hardware like the J1SYS ECG range of pixel controllers. This opens up a lot of possibilities for using 2801/2811 standard type pixels with LOR S3 software.

Welcome to the Future LOR.

Cheers

Aussiephil

[WilliamS]

I think orvilles question got missed. As I will always be using a show computer its easy to jump over it but it is a valid point.

The MP3 controller, what is the overall capacity of that system. As long as the SD card can handle the data and music files that easy, but how do you output from that into a 1.31 univerese so that someone who likes thier display isolated to do its own thing outside?

[edvas69]

GoofyGuy wrote:

I think orvilles question got missed. As I will always be using a show computer its easy to jump over it but it is a valid point.

The MP3 controller, what is the overall capacity of that system. As long as the SD card can handle the data and music files that easy, but how do you output from that into a 1.31 univerese so that someone who likes thier display isolated to do its own thing outside?

I didnt answer this question in the hope that LOR would have better information on this.

But i do remember a thread going back a few months ago that did mention this and that LOR are working on a new standalone MP3 controller that will be able to handle the increased data rates required. I havent seen anything since on this subject.

[Guest wbottomley]

If LOR would have teamed up with Ed Bryson (j1sys) 2-3 years ago, these tools would be far more advanced than they are today.

Jim St.John has done great work as well. Especially with the incorporation of the ELOR for users that will not upgrade to S3. This single device along with the E68X series are pixel perfect.

[zman]

LOR et al. Thanks for the info and update.

For those considering any RGB implementation, if they are following any of the other forums, the message is clear. Older PC's are not going to cut it.

That is part of the decision process an individual has to make, and we can't hold LOR liable for giving us what we want.

However, for those that are looking at this, it would be good if LOR could provide some sort of harware recommendation to support the advanced networks so users can make plans and allocate budget, and users have to understand, those are recommendations, not definitive configurations.

[Charles Belcher]

edvas69 wrote:

Charles Belcher wrote:

John,

You said in your first post...

" You can test using any hardware that supports the standard."

Does this include wireless transmission?

Charles

Wireless transmission of E1.31 is not recommended mainly due to the fact that E1.31 has very limited error correction and this may allow packets of data to arrive out of order or not at all which may have adverse effects on the display.

You can still use E1.31 over a wireless network, but just be aware that you may get some odd effects depending on how good of a wireless connection you have

Where, in the scale of E1.31 do you suggest problems will occur with wireless transmission? In lightweight testing, I have not had problems. lightweight=1 universe

Charles

[edvas69]

Charles Belcher wrote:

edvas69 wrote:

Charles Belcher wrote:

John,

You said in your first post...

" You can test using any hardware that supports the standard."

Does this include wireless transmission?

Charles

Wireless transmission of E1.31 is not recommended mainly due to the fact that E1.31 has very limited error correction and this may allow packets of data to arrive out of order or not at all which may have adverse effects on the display.

You can still use E1.31 over a wireless network, but just be aware that you may get some odd effects depending on how good of a wireless connection you have

Where, in the scale of E1.31 do you suggest problems will occur with wireless transmission? In lightweight testing, I have not had problems. lightweight=1 universe

Charles

This is dependant on many factors, the distance used, the hardware used, the network traffic ect. With minimal error correction there may be issues, but in your testing it is OK, this may not be the case when the show is actually running with all that added electrical interferance that is generated. Using wireless with e1.31 may work perfectly for you and then others may have issues. Having a light load, a short transmission distance and a direct line of sight will definetly help. So again its one of those things that you need to test to ensure it will work for you. But it may not work for everyone all the time.

[aussiephil]

Charles Belcher wrote:

edvas69 wrote:

Charles Belcher wrote:

John,

You said in your first post...

" You can test using any hardware that supports the standard."

Does this include wireless transmission?

Charles

Wireless transmission of E1.31 is not recommended mainly due to the fact that E1.31 has very limited error correction and this may allow packets of data to arrive out of order or not at all which may have adverse effects on the display.

You can still use E1.31 over a wireless network, but just be aware that you may get some odd effects depending on how good of a wireless connection you have

Where, in the scale of E1.31 do you suggest problems will occur with wireless transmission? In lightweight testing, I have not had problems. lightweight=1 universe

Charles

Charles,

Some hopefully valid technical info that may also be useful.

sACN (E131) is transported over the IP network using UDP, UDP is not reliable—messages may be lost or delivered out of order. Now for a lot of stuff we do with sequencing software that is sending constant updates often at 20-40fps the lose of a UDP packet or two would likely not even be notice, HOWEVER, if your sequencing software of choice was to send an update only on change, and that packet was lost or received out of order you would get unpredictable results.

The update on change only is completely valid and these is no requirement for constant updates.

Now the out of order reception can occur on both wired and wireless networks but the likelyhood of the occurence increases with wireless transmission by the very nature of wireless.

If I was using say Vixen that sends all data every time period say 25ms then wireless is quite valid as the refresh is high. Other software may not be as good at hiding the issues.

Like everything you personal mileage will vary.

Cheers

Phil

[Charles Belcher]

Phil/Eddie,

Yes, yes and yes. I was hoping for an official statement/answer from LORJohn.

And, although LORJohn stated he would not identify the max limit of S3, I think LOR should do this. I don't know why it should be a secret.

Just because "Mileage may vary" can most certainly be true;
there are also many "if/then" situations which could be identified.

Charles


aussiephil wrote:

Charles Belcher wrote:

edvas69 wrote:

Charles Belcher wrote:

John,

You said in your first post...

" You can test using any hardware that supports the standard."

Does this include wireless transmission?

Charles

Wireless transmission of E1.31 is not recommended mainly due to the fact that E1.31 has very limited error correction and this may allow packets of data to arrive out of order or not at all which may have adverse effects on the display.

You can still use E1.31 over a wireless network, but just be aware that you may get some odd effects depending on how good of a wireless connection you have

Where, in the scale of E1.31 do you suggest problems will occur with wireless transmission? In lightweight testing, I have not had problems. lightweight=1 universe

Charles

Charles,

Some hopefully valid technical info that may also be useful.

sACN (E131) is transported over the IP network using UDP, UDP is not reliable—messages may be lost or delivered out of order. Now for a lot of stuff we do with sequencing software that is sending constant updates often at 20-40fps the lose of a UDP packet or two would likely not even be notice, HOWEVER, if your sequencing software of choice was to send an update only on change, and that packet was lost or received out of order you would get unpredictable results.

The update on change only is completely valid and these is no requirement for constant updates.

Now the out of order reception can occur on both wired and wireless networks but the likelyhood of the occurence increases with wireless transmission by the very nature of wireless.

If I was using say Vixen that sends all data every time period say 25ms then wireless is quite valid as the refresh is high. Other software may not be as good at hiding the issues.

Like everything you personal mileage will vary.

Cheers

Phil

[Daryl Hurd]

I just want to say Thanks to the LOR people involved in doing E1.31.
I know how hard it is to program new features into software and it sometimes seems like a thankless task.

Also I want to give kudos to everyone there for posting some information on this board. By just having this little bit of insight, it is changing my plans over the next couple of months.
-Daryl

[LightORamaJohn]

Charles Belcher wrote:

... I was hoping for an official statement/answer from LORJohn.

And, although LORJohn stated he would not identify the max limit of S3, I think LOR should do this. I don't know why it should be a secret....

Just because "Mileage may vary" can most certainly be true;
there are also many "if/then" situations which could be identified.

Charles

This sub-thread concerned E1.31 over wifi. The problem here is that E1.31 uses UDP to transport the packets. There is no guarantee of delivery in order or even of delivery at all. In the worst case, you turn off pixels and have an out-of-order event. This would result in a 25ms flash of LED devices. I use 25ms as the default strobe interval in our floods. You can easily see a 10ms flash. Since we don't do update only DMX, it would correct after the flash. In practice, I think this would not be a problem, because if it were, there would be a TCPIP version E1.31 by now.

As far as limits in S3, we're not trying to keep this secret it's just that there are so many variables that affect the limit we can't say something like "our car goes 180mph." In our universe a legitimate question would be "well, what about on Mars?" We tested with 10,000 channels and had no problems. I mentioned in other threads that we see limits in the software and we are working on those. We also have a new product that has been designed but is scheduled for next year where the limit on the number of channels in a show is not limited. It's a different way of approaching the problem and works with LOR or DMX networks.