Has anyone tested Retro-8 from MovieStuff yet?

[Tina]

Hi Roger. Yeah, it was like 5 or 6 days between when I posted and when it showed up on the board. I had asked the mod for the post to be deleted once we got in touch with you but it went through anyway. Sorry about that! Anyway, the unit came today and we are looking forward to setting it up over the weekend!

[prtprt]

A question to Roger:

When I look at the photo of Retro-8 and compare it to a photo of for example WorkPrinter XP, if I understand it correctly I see that the camera is looking at the 8mm film from different sides of the film.

For WorkPrinter XP the MoviesStuff information text is:

And unlike typical projectors or other 1CCD telecine units that crop the image and shoot off the softer, base side of the film, the Video WorkPrinter-XP lets you use a high quality 3CCD camera to image off the sharper, emulsion side of the film...

Is the Retro-8 camera shooting off the softer base side of the film? Or maybe there is not that much different if you compare the sides? Or maybe I misunderstand where the camera is placed on the Retro-8...

Charlie

[MovieStuff]

Unlike all our previous units, the Retro series does scan through the base side of the film. I would prefer the emulsion side but decided that we could ensure critical focus on all films if the emulsion were kept the same distance from the sensor regardless of film thickness. Also, by having the emulsion side down, only the edges of the base side is the surface that rides the rollers during transfer and high speed rewind. So it was a design decision formulated around customers not having to focus and also to protect the emulsion area as best as possible.

Roger

[prtprt]

Thank you, Roger.

Charlie

[carllooper]

Unlike all our previous units, the Retro series does scan through the base side of the film. I would prefer the emulsion side but decided that we could ensure critical focus on all films if the emulsion were kept the same distance from the sensor regardless of film thickness. Also, by having the emulsion side down, only the edges of the base side is the surface that rides the rollers during transfer and high speed rewind. So it was a design decision formulated around customers not having to focus and also to protect the emulsion area as best as possible.

Roger

I don't quite follow this description.

If the Retro is scanning through the base side doesn't that mean the base is sitting between the emulsion and the sensor - which in and of itself is not a great issue if the lens is focused on the emulsion. But if the film is registered in terms of the base, ie. "the base side is the surface that rides the rollers during transfer" how does this setup ensure the emulsion is kept "the same distance from the sensor regardless of film thickness"? The way the setup is described suggests that what would remain a constant is not the sensor to emulsion distance, but the sensor to base distance. But if this were the case then different film thicknesses would throw out focus.

Or am I misreading the setup?

EDIT: Ok - just had a look at the unit (which I should have done first) and I get the idea now. The emulsion side of the film engages the gate (and is therefore always the same distance to camera). And by rollers is meant those rollers either side of the gate - which engage the opposite side of the film: the base.

So what would have been ideal (but not that necessary) is to have the camera underneath and the light above - but then the controls would need to be put elsewhere which would break the otherwise compact design.

C

[MovieStuff]

...Ok - just had a look at the unit (which I should have done first) and I get the idea now. The emulsion side of the film engages the gate (and is therefore always the same distance to camera). And by rollers is meant those rollers either side of the gate - which engage the opposite side of the film: the base.

So what would have been ideal (but not that necessary) is to have the camera underneath and the light above - but then the controls would need to be put elsewhere which would break the otherwise compact design.

Exactly. A lot goes into sorting out the ergonomics of the design, production methods, number of parts, size of the unit relative to shipping costs, etc and weighing that against perceived differences in sharpness of emulsion vs. base side scanning. Ultimately, the design we settled on had base side scanning for all the reasons already stated plus the considerations just outlined. Everything is a compromise when dealing with a commercial product as opposed to a custom built item where you don't have to repeat success relative to budget over and over in a predictable fashion.

Roger

[carllooper]

...Ok - just had a look at the unit (which I should have done first) and I get the idea now. The emulsion side of the film engages the gate (and is therefore always the same distance to camera). And by rollers is meant those rollers either side of the gate - which engage the opposite side of the film: the base.

So what would have been ideal (but not that necessary) is to have the camera underneath and the light above - but then the controls would need to be put elsewhere which would break the otherwise compact design.

Exactly. A lot goes into sorting out the ergonomics of the design, production methods, number of parts, size of the unit relative to shipping costs, etc and weighing that against perceived differences in sharpness of emulsion vs. base side scanning. Ultimately, the design we settled on had base side scanning for all the reasons already stated plus the considerations just outlined. Everything is a compromise when dealing with a commercial product as opposed to a custom built item where you don't have to repeat success relative to budget over and over in a predictable fashion.

Roger

Yes, the way I understand it the sharpness of the image shouldn't really be affected by which side you scan through because, unlike contact printing (where the image would have an opportunity to spread out and become soft before it reached the print stock), there is a lens involved in scanning (in optical printing), which is focused on the emulsion. In other words, while the base scatters the image the lens only sees those light rays (so to speak) which haven't scattered. The same thing happens when photographing anything - surfaces scatter light in all directions but the lens only sees those "rays" which the lens has selected (ie. those going from the surface to the lens/sensor). And as for any damage to the base (scratches) these will transfer from which ever side you image the emulsion. The only thing the base scatter should do is produce a minor decease in contrast (ie. fog) but that's so very easy to remedy in digital post. One just tweaks the contrast on the result a touch more than one otherwise might.

In other words, in addition to the concept of "no perceivable difference" in sharpness, I'd suggest (propose/conjecture/entertain) that with a correctly focused lens there should also be no theoretical difference in sharpness.

C

[BLXRX]

Roger

Are there any provisions or hardware allowances for a future upgrade of the camera/sensor on the Retro 8/16 machines? Thanks.

[MovieStuff]

Roger

Are there any provisions or hardware allowances for a future upgrade of the camera/sensor on the Retro 8/16 machines? Thanks.

The design is fairly modular with that intent. But it will be a while before we go down that road. We need to build up some history with the unit and then we'll look at changes or additions to the design.

Roger

[sjpro]

I just did a test comparing a Retro-8 scan to a Millennium II scan from pro8mm with negative stock. I did basic color correction in Media Composer to the Retro-8 footage. I wasn't trying to match the footage. I think it holds up well even though it's a bit grainier. I had to scan the film 3 times to get the best results since it was confusing at first for me to scan negative because it's inverse so if something is over exposed you have to turn the exposure up instead of down. I had enhance and grain reduction selected in the retroscan software. I'm actually happy that you can see the stenciled letters on the white van better with the retro-8 (01:18) than with the MII (00:54).

https://vimeo.com/82026001

[MovieStuff]

I just did a test comparing a Retro-8 scan to a Millennium II scan from pro8mm with negative stock. I did basic color correction in Media Composer to the Retro-8 footage. I wasn't trying to match the footage. I think it holds up well even though it's a bit grainier. I had to scan the film 3 times to get the best results since it was confusing at first for me to scan negative because it's inverse so if something is over exposed you have to turn the exposure up instead of down. I had enhance and grain reduction selected in the retroscan software. I'm actually happy that you can see the stenciled letters on the white van better with the retro-8 (01:18) than with the MII (00:54).

https://vimeo.com/82026001

Very cool! Thanks for doing the side by side. I've been seeing a number of people using Neat Video for grain reduction with the Retro-8 and it really seems to work well. The grain reduction we use in the RetroScan-HD software is really just a reduction is sharpening because we didn't want to "preprocess" the footage in case people wanted to use an actual grain reduction program later. That's one of the things the "big iron" transfer units have over the Retro-8, which is sophisticated real time grain reduction. Of course, there's a price you pay for that privilege. That said, I'd be curious to see this same footage run through the Neat Video tool and a bit more color correction to see how it compares to the MII. Thanks again.

Roger

[Tscan]

Looks good, and i think it can look a lot closer to the pro scan with a little more gamma and color correction, as well as some grain reduction. I've been experimenting with the Vision3 and Retro8 a lot lately... the amount of head room and control you have over your image in NLE is incredible. It's a whole new world from working with reversal transfers but it takes a little getting used to. Neat Video is very effective and I think you can achieve the same amount of grain reduction with it as well. Until recently I was taking a shot with a Kodak grey card on each film stock for my noise profile. But I shot my last batch using a Neat Video calibrated color target, which i'll post a link to. You print it out on pure white paper and use it for your opening shot. Then you use the center green square (a mix of all the colors) to make your noise profile. From there you also have 3 levels of sharpening you can apply if needed.
http://www.neatimage.com/im/target/NICa ... 0color.png

[mudfly9]

For me, the Retro-8 scan stands up in its own right as a more retro looking alternative rather than simply being not quite as good. Nice work!

[MovieStuff]

This just showed up on YouTube. Kinda neat.

http://www.youtube.com/watch?v=kCyF5ksC ... ata_player

Roger

[JeremyC]

This just showed up on YouTube. Kinda neat.

http://www.youtube.com/watch?v=kCyF5ksC ... ata_player

Roger

Very good pictures.