Film-Tech Forum ARCHIVE: Master Image.

This is topic Master Image. in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=000431

Posted by Frederick Lanoy (Member # 5416) on 09-03-2009, 07:35 AM:

Hello everybody,

I'm a french projectionnist who also works for a website dedicated to digital cinema : www.manice.org.

I'm looking for informations about master image's 3 D system. I had some by their technicians but i would like to have your feedback as professionnals. Indeed, i wrote this article about 3 D : http://www.manice.org/rubrique_glos.php?id_rubrique=59#515

I would like to write a special part for master image like i did for Xpand, Dolby, IMAX...

Thanks for your help,

Frédérick.

Posted by Todd McCracken (Member # 4718) on 09-04-2009, 01:21 PM:

We have one in our lab, its a nice system indeed. Coolest thing about it is that you can move it from house to house as needed. That way you are not stuck with a feature in a big house after its initial draw because your feature is anchored to a non-movable 3-d system.

Posted by Mark Gulbrandsen (Member # 72) on 09-05-2009, 12:04 PM:

Master Image units do give nice results and the movable aspect is an excellent bonus.

However there are three downsides to them that come to mind.

The first is that the filters tend to attract dirt and grime from the air and the fast 4320 RPM rotational speed certainly assists in this factor. Some locations I know that have them have to clean the filter wheel about every 10 days. Once the filters get the grime covering the polarizing ceases to function. It has to be done with a special Chamois which is luckily inexpensive and available at Wal-Mart. The average booth operator will probably end up damaging the filters in the long haul. The good thing though is the filter wheel is not that expensive!

The second problem is that the control logic is somewhat primative. There are several operational modes one of which stops the wheel completely when the projector stops the pulse at the GPIO. In this mode it is possible to burn through the wheel with light from the projector hitting it. The other mode takes the wheel to half speed when the pulse stops. IMHO there should be just the half speed mode so that when the lamp is lit the wheel spins and lessens the chance of a burn through of the filters.

Thirdly, The elevation (up/down) aspect really sucks. It is beyond slow and the prospect for a typical booth operator not getting it set back to the correct height is consistantly possible. The better solution would be to place the unit on a short track system and slide it out of the way when showing 2-D. A simple drop pin would position it in the in or of operating position... something a typical booth operator would be more likely accomplish correctly.

The unit is extremely well built and very heavy for what it is.

Mark

Posted by Brendan Penny (Member # 5041) on 09-06-2009, 05:36 PM:

I agree with you there Mark. We are looking at getting some changes done to have a faster gearbox on the retractable drive. If we could get it so it automatically retracts depending on sync that would be awesome.

It really needs to be able to retract in around 8 to 10 seconds.

Other than that you can't fault the 3D quality compared to other systems. No license fees too. Can't complain there!

Posted by Mark Gulbrandsen (Member # 72) on 09-06-2009, 11:12 PM:

Brendan,

In some installations it can't retract low enough to get out of the light path... mainly installations with alot of down tilt. Already ran into this once. I feel a track is the better way to go. Use the gear box to move it along the track if you like. Easily done!! The retractor part is stupid... I also find it hard to believe that no one is buildng bootleg versions. It's so dam simple you could do it for about 2 grand...

Mark

Posted by Steve Guttag (Member # 268) on 09-07-2009, 06:38 AM:

Perhaps an industry for you to jump into? I mean, even if you were to make 5X profit (10,000) per system, it be, by far, the cheapest 3D solution out there.

Steve

Posted by Julio Roberto (Member # 4976) on 09-07-2009, 11:20 AM:

I'd buy 2 clones for $10k each. There is also a continous (small, but significant) revenue potential supplying cheap glasses (say $0.50) for life to customers of such systems.

Posted by Mark Gulbrandsen (Member # 72) on 09-07-2009, 06:16 PM:

I'm way too busy to have time for that. Wish I could take the time though as 30K for that unit is a bit on the ludricous side considering what it is.

Mark

Posted by Brendan Penny (Member # 5041) on 09-07-2009, 08:30 PM:

quote: Mark Gulbrandsen
In some installations it can't retract low enough to get out of the light path... mainly installations with alot of down tilt. Already ran into this once.

That's a good point actually. I haven't come across that issue yet but I can imagine that could be a problem.

For some scenarios the up/down retraction would work and it would cost neither money or time to change the current gearbox ratio in the unit. That's simple solution number 1 which may help some people.

In terms of a slider unit. Sounds good but I just don't have time to work on these science projects anymore!

Posted by System Notices (Member # 2357) on 01-16-2011, 07:39 AM:

It has been 495 days since the last post.

Posted by Marco Giustini (Member # 4544) on 01-16-2011, 07:39 AM:

I was doing some math about MI and 3D and there is someting that puzzles me.

I was reading the MI install guide and I learnt that you need to frame the light beam within the disk segment. But if the picture is so tightly fitted inside the segment, shouldn't you have a very long dark time, since the segment is changing for most of the time?

So I did some calculations: 144 fps means that each frame lasts for 6.94ms. MI sets - on the user guide - a Dark time of 1.157ms. I guess - from pictures - that the disk has 16 segments. From the manuale in 3D mode the disk spins at 4320rpm, 72 revolutions per second. This way 1152 segments are shown in front of the lens for each second.

?? We have 144 frames and 1152 segments, x8 the frame rate?

It would make sense if the filter was split in two (72 x 2 = 144) but if this was the case, why those lines to separate the sectors?

Does anybody know how it works?

Posted by Carsten Kurz (Member # 5396) on 01-16-2011, 10:29 AM:

I think adjacent segments do NOT all change L/R polarization, but only 're-align' segment angle due to the wheel rotation. That means, multiple adjacent segments share the same L/R polarization. MI does tripleflash, so 24fps and 4320 UpM indeed means that the wheel is split into L/R halves. But due to the varying rotational angles, they have to glue segments together that realign the filter segment angle. So each segment is cut and rotated slightly against it's neighbour, the full segment sequence hence would be L/L/L/L/L/L/L/L/R/R/R/R/R/R/R/R.
And I guess they simply don't do blanking between adjacent L or R segments.

RealD has a static filter alignment and thus works with a full format filter, be it ZScreen or the Sony XLS.

Yes, if you think this through, you will come to the conclusion that this MI system has some strange implications with its 8* jittering polarization realignment, and you wonder how it is possible for it to deliver a decent 3D experience at all.
Seems that the human eye and brain is just to easy to be cheated.

- Carsten

Posted by Steve Guttag (Member # 268) on 01-16-2011, 10:50 AM:

Its 3D...the human brain/eye has just learned to expect less!

It does work surprisingly well. Much better than Real-D, in my experience and it is more light efficient too.

-Steve

Posted by Marco Giustini (Member # 4544) on 01-16-2011, 11:53 AM:

Indeed, our Brain is less smart than I thought [Smile]

Thanks for the explanation!

Posted by Marco Giustini (Member # 4544) on 01-18-2011, 12:32 PM:

To be pedantic, the segments are actually 12. A filter picture can be seen on the replacement guide in the warehouse. [Smile]

Posted by Carsten Kurz (Member # 5396) on 01-20-2011, 09:23 AM:

Oh, yes, interesting pictures and statements in these new documents, at least for those that have never seen a MI system in real life. Also interesting, the official statements concerning Mk1 and Mk2 systems.

- Carsten