Reflecta RPS 7200 - any opinions?

[xtolsniffer]

And now via Silverfast - adjusted in the software (histogram) but no ICE or sharpening. First 3200 dpi, then 7600 dpi.

 

[chuck94022]

This is interesting. I'm having a hard time telling whether you are actually getting more information other than DMax out of the Reflecta over the Epson.

 

[xtolsniffer]

This is interesting. I'm having a hard time telling whether you are actually getting more information other than DMax out of the Reflecta over the Epson.

I have to agree, at least on that image. To be honest, I'm not sure how much more resolution there is to be had in that one, it was hand-held, so any more detail is going to be blurred anyway. I need to find a transparency that's got a good level of detail from a slow film from a tripod shot. And I need to do it before the 30 day trial of Silverfast expires! The thing that I've been most impressed by is actually just how good the Epson is. It does get panned for 35mm work (online at least) but I've done shots from 35mm scans up to 12"x16" and they are pretty good at normal viewing distances. What's held it back for me is not the resolution but keeping detail in the shadows while not blowing highlights despite almost infinite amounts of software fiddling. The tests continue....

 

[pellicle]

Blown highlights were my nemesis with the V700, even when very carefully adjusted. When I've got time to do a proper side by side comparison I'll post samples.

this makes me think that you are not driving the scanner properly. There should be no blown highlights on the Epsons. So if you don't mind me asking:
* did you mean blown highlights on colour negative, slide, or black and white negative?
* have you tried using the epson scan software? -> if so are you setting your black and white points yourself or letting it do it?

have you read my blog posts on setting points for the Epson scaners?
(just in case it is here in my view ...: black and white neg scanning)

be curious to hear back about your highlights

 

[xtolsniffer]

OK, new scans. Both Epson V700 and Reflecta RPS7200. Image is from Provia, tripod mounted camera (Nikon F4, 105mm macro), probably a tad underexposed. Scans are at 3600 dpi, within the maximum of both units. I chose that value as it's half the maximum theoretical resolution of the Refelcta, and also it's plenty big enough! Scans are done via Silverfast (licensed copy for the Espon, trial for the Reflecta, hence the watermark). No manipulation was done other than a very slight adjustment to the histogram. Here are the full images, resized for here. First the Epson then the Reflecta.

 

[xtolsniffer]

And now detail taken from the original .tiff file, again first the Epson, then the Reflecta.

 

[xtolsniffer]

My view of this, after looking at the full size .tiff files at 100% is that there is a tiny bit more resolution coming out of the Reflecta rather than the Epson. Hardly noticeable, but it is there, just. Superficially, the Epsons look better scans, more contrasty, punchier colours, but the Reflecta scans are easier to post-process. They sharpen better, and you have more scope for contrast adjustment. This means you do have to spend more time sorting them out, but I guess you could put most of it into a batch process, I do that in Capture NX2. The bottom line is that you have a £500 Epson scanner that does 35mm, MF and bigger, and also reflected scans, vs £500 for a scanner that does 35mm only (plus the cash for Silverfast/Vuescan). The Reflecta is no doubt easier and faster to use, you just load the film in, and away it goes. The Epson has fiddly holders and is a nightmare for dust, and on that note, the Refelcta is much better. I've spent a lot of time setting up the Epson to squeeze the most out of a transparency, I'm not there yet with the Reflecta. It's certainly no magic bullet, but it means that I can happily carry on using film and getting it onto digital medium. Having compared my scanned slides with images from my Nikon D700, the D700 wins in terms of resolution, no contest. But to be honest, it's dull, no fun to use. I like my F4, I like being slow, I like taking only a few images of each subject and thinking about it more. Given that a 3600 dpi scan, and a Nikon D700 will print to 12"x16" easily, how much bigger would I ever need to print?

 

[pellicle]

My view of this, after looking at the full size .tiff files at 100% is that there is a tiny bit more resolution coming out of the Reflecta rather than the Epson. Hardly noticeable, but it is there, just.

but will it print?

I've done some examinations of pixels in the print on inkjets and on laser RA-4 type prints and found that there is always a bit of pixel blurring (say, ink splatter or light diffusion in the paper partially activating the adjacent space on the paper in RA4)

Ive embedded text which has hard edges in the image and printed to see what effect it has:

If you feel like reading the article (sorry, its long) its here in my view ...: digital vs (35mm) film, scans screens and prints

I often write my blog so that I can explore an idea and be forced to write it sensibly AND also get a reference back to something so I don't have to dig through archives of (say) this place to find what I wrote.

PS: I'd need a bigger sample of the image to see any difference. Perhaps its "just" there .. perhaps you just wanted to see it? I saw something but can't be sure its extra details. I've rescanned the same image on scanners (ls-4k and ls-40 nikons, various Epsons) and had slightly different results each time. Could easily be that you need to scan the image at least twice on each scanner and compare averages.

which (to me) means that if you are going to that level of examination you could call it a tie.

 

[xtolsniffer]

Good article, thanks for that. I like the analogy of the cat and the fish tank, you can see there is something in there but you can't get it out. I think the differences are so small, you'd have to look really hard to find them. One day, when I can think of nothing better to do, I might print off a test target, not that is has much to do with the real world, but for a dark winter's night, well why not?

 

[pellicle]

I might print off a test target, not that is has much to do with the real world, but for a dark winter's night, well why not?

excellent idea. I did a lot of this work when I was living in Finland, where the winters nights are not only dark, but bloody long (and cold).

Worth grabbing is a Stouffer step wedge. This little bit of stepped plastic will help you take the mystery out of what the DMAX of your scanner actually is.

in my view ...: Testing LS-4000 with Stouffer Stepwedge

in my view ...: Epson 4990 response testing

in my view ...: Using Epson flatbed scanners for densitometery

 

[I.G.I.]

Perhaps a bit fairer (but more tedious) comparison would be to test each scanner driven by it's native software; try to extract the best TIFF file from it, and then post side by side. I've heard more than a few times that in fact the Epson software is pretty good; and according to some gives better results than the third parties. Reflecta on the other hand is known that works much better with SF than with VS.

 

[Felinik]

Perhaps a bit fairer (but more tedious) comparison would be to test each scanner driven by it's native software; try to extract the best TIFF file from it, and then post side by side. I've heard more than a few times that in fact the Epson software is pretty good; and according to some gives better results than the third parties. Reflecta on the other hand is known that works much better with SF than with VS.

What would that give us?

All that most of us are interested in here is how to achieve the best scans we can, at the least possible cost.....

 

[I.G.I.]

What would that give us?

All that most of us are interested in here is how to achieve the best scans we can, at the least possible cost.....

Exactly... but the way to achieve the best scan is to choose the software that works best with each particular scanner, that's why I proposed to try the optimal driver for each. To judge the output from different scanners using Vuescan seem to me contrary to the common sense.

Cost is not much relevant here because the OP already owns Epson, and respectively has the Epson scan. As for Reflecta, it comes with some kind of crippled version of Silverfast; if one wants full functionality, including saving files in 16 Bit mode, the paid upgrade is inevitable.

 

[Felinik]

Exactly... but the way to achieve the best scan is to choose the software that works best with each particular scanner, that's why I proposed to try the optimal driver for each. To judge the output from different scanners using Vuescan seem to me contrary to the common sense.

Cost is not much relevant here because the OP already owns Epson, and respectively has the Epson scan. As for Reflecta, it comes with some kind of crippled version of Silverfast; if one wants full functionality, including saving files in 16 Bit mode, the paid upgrade is inevitable.

Well, now I don't know a lot about scanner drivers, but in the case of the Reflecta, the software shipped with the scanner is crap, unless you buy the more expensive pack that includes Silverfast. It seems that no matter what scanner you use Silverfast excels in terms of the end result. The thing with the Epson software is that you can achieve good result pretty easy, but afaiu with Silverfast you rock the h*ll out of the Epson's too...

 

[I.G.I.]

That's the catch with the current 35 mm scanners on the market: both Reflecta and Plustek ship with crap/crippled software bundles that need immediate full price upgrade; apparently neither manufacturer cannot be bothered to develop adequate software to ship with the scanners.

If I was in the situation of the original poster I would definitely try the Reflecta with SF (I'm not a SF or VS user so no personal bias here; I scan with Minolta hardware and software)

---

Addendum. My bad, I got somehow carried away with the impression from the earlier samples that the images from Reflecta were only with VS; now I realised the last were from SF... sorry for my lapse...

 

[xtolsniffer]

Well, I've just done a series of tests with a lens test chart, same lens, same distance, same target, same light using a Nikon F4 with Velvia, an F100 with Portra 160 and a D700. I'll scan the material when it gets back from the lab on the Epson and the Reflecta and post results.

 

[I.G.I.]

Velvia/Portra/D700, with Epson and Reflecta parallel scans really tickled my curiosity, can't wait to see your comparisons posted.

 

[xtolsniffer]

Well the Velvia has gone off for processing, so we'll see about that. The Portra will have to wait until I've finished the roll. The D700 images are useful for a check of the lens/chart system. I'll post those as well. At the moment, they look like they resolve about 55 lp/mm on the chart which I think is close to the theoretical maximum for that sensor. I'm sure someone will chip in about lp/mm on a digital sensor, but I'm just going by what I can see on the charts

 

[Felinik]

I've been doing some B&W scan testing here the last days using Vuescan, apparently there's some kind of hickup so when using the IR Clean through Vuescan, everything just freezes. IR Clean works perfectly well with the Cyberview X5 software, and Ed Hamrick is on the case trying to find out what's going on.

Anyone else on Mac (10.7.5) who's got Vuescan, latest Reflecta Firmware+Driver who'd be up for testing scanning (I can send you my .ini file to get the exact settings) and use the IR Clean option to see if it works for you or not?

 

[Rudeofus]

Felinik, I hope you don't try to IR clean regular b&w film, it won't work. Unlike many dyes, Silver is just as impenetrable to IR light as dust or scratches.

 

[Felinik]

Oh really?

No matter what, it makes things crash...

 

[Rudeofus]

Oh really?

Most dyes used in color films (or chromogenic b&w films) are pretty much transparent to IR light to the point that you can use unexposed and developed slide film as capable IR pass filter. When a scanner or the software that drives it uses ICE (or whatever the IR clean mechanism is called), it makes an extra scan pass with IR light. Whatever shows up intransparent in the IR scan is likely dust or a scratch and the affected area must be interpolated from the surrounding image matter by software algorithms. Since Silver is opaque to both visible and IR light, ICE won't work with regular b&w film.

No matter what, it makes things crash...

I agree 100% that the scanner software shouldn't crash under these circumstances and I'm confident that Ed takes care of this problem quickly.

 

[Felinik]

What you are pointing out is interesting, as the same problem appears if I set vuescan to do a multipass scan... Crashing..

Apparently there's something into the "extra pass" thing, at least without knowing anything about how these things works, makes one speculate around this...

 

[Rudeofus]

Apparently there's something into the "extra pass" thing, at least without knowing anything about how these things works, makes one speculate around this...

I use vuescan regularly and without these issues under linux and with my Epson V700, includig multipass and ICE. Given that Reflecta supplies only a BLOB to Ed Hamrick (that's why most likely there is no linux support for vuescan+Reflecta) I foresee a bloody mess when one has to debug this issue. I'm glad I didn't spend money on this scanner and wish you luck that this issue gets resolved soon.

 

[Felinik]

I use vuescan regularly and without these issues under linux and with my Epson V700, includig multipass and ICE. Given that Reflecta supplies only a BLOB to Ed Hamrick (that's why most likely there is no linux support for vuescan+Reflecta) I foresee a bloody mess when one has to debug this issue. I'm glad I didn't spend money on this scanner and wish you luck that this issue gets resolved soon.

Odd thing is that it works with crappy Cyberview X5, so I guess either there's a problem with the reception of the Vuescan calls in the scanner driverlib, or else somethings prolly really effing up.

Anyway, Ed's on the case, and I am just waiting for him to get back to me when he gets his head around it.