Big Panoramic Screen - 6 projectors

Hello,...

Need help, or Advice ?

I do have a job, next october, where we will be making a 6 projector Panoramic Screen... Blend and evrething. Using 6 x 15K projectors times 2 Stacked.

The question is ??? The guys here will be making a Movie of arround 3200 x 480 pixels. Not the exactly number at the moment, since the screen is TBD.
But will be arround that, like 50% or so od the Total pixels of the screen.

The questions ARE:

We will be using Catalyst to do it, with TH2G, Mtron SSDs etc...
Te thing is, better to RUN 1 Catalyst with 2 TH2G, making the 6 screens @ 2400 x 600 each TH2G, that way the SYNK is easy and better.
OR
Use 2 Catalysts SYNK, and 1 TH2G on each one on out 2, and use each Cat to half od the screen. The movie will be the same, but Offset, of course.

Another question, we wanna know if we have Captured with LFG Phoenix SDI cards, if we get almost no DELAY..... We only have Blackmagix Extreme HD cards, and we experience a huge delay on the LIVE capture, when playingback a movie tht size.

Please SHARE your comments.

King regards

Gian

you have to set up and test for yourself.

you may or may not need more than 6 outputs. thats quite small.
blending gets easier the more overlap there is.

there is a lot of projector stuff that needs to be correct-

i have done it myself several times. but it is tricky. and used more than 6 outputs.

playing a movie 3200x480 takes almost all available time on computer. so there is no time left to do live video. and frames get buffered by blackmagic cards.

Quote:

---

Originally Posted by Gian

is, better to RUN 1 Catalyst with 2 TH2G, making the 6 screens @ 2400 x 600 each...
OR
Use 2 Catalysts SYNK, and 1 TH2G on each one

---

Short answer: Use 1 machine and have 1 backup system.

Quote:

---

Originally Posted by Gian

Another question, we wanna know if we have Captured with LFG Phoenix SDI cards, if we get almost no DELAY.....

---

My visual judgement is about 2-3 frames using LFG4 and 3-5 frames using 4 lane Phoenix. However, 4-6 frames on Extreme so you'll see the same issue.

THANKS SourceChild

We do here almost veryday, 4 projection Blending Panoramic screens, but that is about 1 TH2G. On our tests and playing here, we think like you.
But since 2 TH2G are involved, like more opinions.
Thanks again

Regarding the LFG cards, since I need quality, We will not buy LFG cards at the moment, All our Cat sys have Blackmagic Extreme HD cards.
With a Still frame, so no mivies behind looks "OK", but we are planning at the moment in add some Encores Barco in beetween CAT and the projectors.

King Regards

Gian Bortolotti

no- delays on blackmagic are from a totally different source-

blackmagic buffers frames when the computer is doing something else- especially when playing back a large movie at the same time....

and it can buffer up to more than 1/2 second. especially with hdsdi and playing lots of movies or a big movie.

this is totally out of my control- and variable - it changes - i have seen up to 2 seconds of buffering- which is good for video capture to disc- but not so good for doing live input.

thats because its designed to capture video onto disc with no frame drops- and the time delays through the buffer are not important for their normal users.
Latency is not a design problem for video capture to disc.

---

LFG and phoenix cards dont buffer like this-

delay is constant - and is done by me and graphics card-

Quote:

---

Originally Posted by SourceChild

My visual judgement is about 2-3 frames using LFG4 and 3-5 frames using 4 lane Phoenix. However, 4-6 frames on Extreme so you'll see the same issue.

---

Quote:

---

Originally Posted by Gian

Regarding the LFG cards, since I need quality, We will not buy LFG cards at the moment, All our Cat sys have Blackmagic Extreme HD cards.
With a Still frame, so no mivies behind looks "OK", but we are planning at the moment in add some Encores Barco in beetween CAT and the projectors.

King Regards

Gian Bortolotti

---

You should try phoenix cards for sdi- ( LFG is composite ) ( LFG and phoenix are not the same - ) LFG is composite- Phoenix is multichannel hd/sdi.

i do not recommend decklink cards anymore because of the way they buffer frames.

Quote:

---

Originally Posted by samsc

LFG and phoenix cards dont buffer like this-

delay is constant - and is done by me and graphics card-

---

What do you mean "done by you?" I am aware of the aspects of the graphics card delays but how does Catalyst delay?

Quote:

---

Originally Posted by SourceChild

What do you mean "done by you?" I am aware of the aspects of the graphics card delays but how does Catalyst delay?

---

every IO subsystem of a computer typically has to be double of triple buffered ( or more ) ( in computers terms its called a queue or fifo )- depending on the nature of the source and the output-

its a standard engineering problem- with systems that operate at subtely different speeds.

nothing is exactly 25fps or 30 or whatever.

anything that connects to anything else needs to be buffered-

no 2 computer systems ever have exact clocks- usually 60parts per million.

with the decklink drivers- they buffer frames up to 2 secs- so they have a buffer which can hold say 60 frames - this is what causes the "huge" delays with decklink. decklink is designed to get sdi onto a disc which no frame drops- so they have big long buffers of frames in case of slow disc speeds. Decklink uses a long FIFO buffer.

with phoenix card- i double buffer the input into my software- and the graphics card is double buffered - so 2-4 frames - i dont need to write to disc.

---

its a classic engineering problem...
http://en.wikipedia.org/wiki/Double_buffering
http://en.wikipedia.org/wiki/FIFO_(computing)

Quote:

---

The easiest way to explain how double buffering works is to take a real world example. It is a nice sunny day and you have decided to get the paddling pool out, only you can not find your garden hose. You'll have to fill the pool with buckets. So you fill one bucket (or buffer) from the tap, turn the tap off, walk over to the pool, pour the water in, walk back to the tap to repeat the exercise. This is analogous to single buffering. The tap has to be turned off while you "process" the bucket of water.
Now consider how you would do it if you had two buckets. You would fill the first bucket and then swap the second in under the running tap. You then have the length of time it takes for the second bucket to fill in order to empty the first into the paddling pool. When you return you can simply swap the buckets so that the first is now filling again, during which time you can empty the second into the pool. This can be repeated until the pool is full. It is clear to see that this technique will fill the pool far faster as there is much less time spent waiting, doing nothing, while buckets fill. This is analogous to double buffering. The tap can be on all the time and does not have to wait whilst the processing is done.
In computer science the situation of having a running tap that cannot be, or should not be, turned off is common (such as a stream of audio). Also, computers typically prefer to deal with chunks of data rather than streams. In such situations double buffering is often employed.

---

Quote:

---

Originally Posted by samsc

every IO subsystem of a computer typically has to be double of triple buffered...

...with phoenix card- i double buffer the input into my software- and the graphics card is double buffered - so 2-4 frames - i dont need to write to disc...

---

I gotcha now. I was thinking you meant you defined a dynamic buffer function into the software.

Do you mean you're simply using the standard C++ Quicktime I/O API buffer?

Quote:

---

Originally Posted by SourceChild

I gotcha now. I was thinking you meant you defined a dynamic buffer function into the software.

Do you mean you're simply using the standard C++ Quicktime I/O API buffer?

---

there are many types of buffers. i use many different types of buffers and queues.
i use at least 4 different types of fifo - they are required to do simple things... its quite complex getting all this stuff onto the screen from different sources.... at different frame rates ( and now everyone uses triple heads and double heads, and all that stuff which totally confuses the graphics card... )

i dont use any c++, i dont use quicktime buffering directly -

but quicktime has its own decompression buffering strategies for multi-processing handling and sending out jobs to decompress frames.

i posted images from a 360 panorama i did with improve in another thread.

Image captured via remote desktop - show all 8 output screens in the remote desktop screen capture-
you can see all 8 screens on multiple computers in remote deskop.

http://chaldee1.gotadsl.co.uk/~richa...hlight=improve

Quote:

---

Originally Posted by Gian

Hello,...

Need help, or Advice ?

I do have a job, next october, where we will be making a 6 projector Panoramic Screen... Blend and evrething. Using 6 x 15K projectors times 2 Stacked.

Gian

---

Thanks

ok I have the question. if I have connected all catalyst and controls all with the help HOG 3 and each of catalyst are connected to the projector. it whether creature one large file of the video or a few files with the cut file??

Quote:

---

Originally Posted by Wojtek.koperski

ok I have the question. if I have connected all catalyst and controls all with the help HOG 3 and each of catalyst are connected to the projector. it whether creature one large file of the video or a few files with the cut file??

---

that depends on what the overall size of the projection is-

there is a tradeoff between image size and number of layers that can be played-

generally images are cut up - usually automated in something like after effects.

ok so i have to preepear i large file and the cut him in 3 same pises in AE (or another program).