Epson RD-1 same sensor as the Nikon D100?

[Benjamin Marks]

I remember dimly that the Epson RD-1 had the same imaging chip in it as the Nikon D100. I find the IQ of my old Epson files to be very good. No comparison at poster-size with the Leica M9 of course . . . the resolution wars are long lost for those older chips. But. But. But. This week I got the itch to revisit the chip and, ya' know, the D100 costs about $100 from KEH . . .Yeah, you know how this story ends. With smiles. Every time I post about this, I get at least one response about how the experience of working with a rangefinder is qualitatively different from working with an SLR. I don't dispute that, but from time to time I am really more interested in the qualities a picture has than the qualities the gear has. And, I know there is endless debate about this, but for me we have been technically "there" ever since the Nikon D3 came out. Technical improvements over the D3 or the M9 are really gravy to me. So this morning I got out the D100 and started roaming the house.

Tech/specs: Nikon D100, 35/2 AF-Nikkor (no D), B&W conversion in Silver EFX Pro to Tri-x (sort of), EXIF data ses f3.5,1/8 second hand held, but I am usually not that steady, so it is suspect:

I have to say I am very happy with the output. Now, of course, there have been many advances in how cameras work since the heyday of the D100. Current cameras can operate well in many situations that older cameras can't, throughput to storage media is an order of magnitude faster, image stabilization blah-blah-blah. BUT if your goal is a certain style of output rather than being wedded to a particular tool, I would posit that good results can be obtained even from older, currently "obsolete" tech.

Here's more updated (but still a generation back) tech. Snaps from a Fuji X-Pro 1.

"Better"? Sure. But I think that I am going to try a series of pic with the D100 with the goal of duplicating the IQ from the Epson (which I sold to buy an M8, which in turn I sold to buy an M9 . . . which in turn made me look at my 401(k) and realize that I have to get off the upgrade merry-go-round). 🙁

[Edit: The Nikon D100 was set at its base ISO 200 and some grain added by Nik Silver EFX Pro. The Fuji was set at ISO 800 -- which really show how far these chips have come in the past 10 years.]

 

[willie_901]

I believe it is the same CCD photodiode array/electronics circuitry as the D100.

The IR filter layer, micro-lens and, or color-filter array assemblies may be different.

The approximately two-stop difference you estimate is consistent with the difference in read noise statistics between the two sensor assemblies (data). Incidentally the noise level difference is consistent with using a 12 bit ADC vs a 14 Bit ADC.

At ISO 800, the X-Pro 2 has at least one stop less noise.

 

[p.giannakis]

I have a canon 10D from the same vintage the only digital camera I own. I have been very happy with the picture quality, i never considered upgrading to a newer one. The D100 looks good too.

 

[B-9]

I thought it was the D70 but my memory is not the best.

Good news though, D100's are sooo much cheaper than RD1's

 

[zane0777]

I think the D100, D70, and Pentax DS all used the same Sony 6 MP CCD sensor. I still enjoy the DS's images, though IQ suffers a bit over ISO 800.

 

[BLKRCAT]

Very much prefer the top image over the 2 xpro images.

Are those OOC black and whites? I kind of feel like they are with the amount of noise reduction and sharpening that they appear to have on them. It's such an off putting look for me. Too pastel like.

 

[Benjamin Marks]

Well, on the bottom two pix I pulled a bit of a bone-headed move. Then again, I was just fooling around, and I hadn't really gotten out the Fuji X-Pro 1 in several months. I set the film choice to B&W and the file choice to RAW. Duh. Also ISO to 800. Who knows what that combination does in-camera? The RAW files read as color, of course, which I then converted to B&W in PS5 using Silver EFX Pro, probably on the Acros 100 setting. Buried in there too is a conversion to 16-bit RGB, and then to Greyscale and a down rez'ing and conversion to jpg so I could post it. I say "probably, " because with the number of options for getting to this kind of output unless I do things only one way all the time or take careful notes, there is pretty much no way to reconstruct the steps that go into the formation of a particular image.

I have an annual project in addition to whatever else I have going on in a particular year: I produce an annual year book of my kids with the best snapshots from each month. There are some nephews in there too. Past years have been a mix of color and B&W, and I am toying with the idea of going all B&W this year. Actually, I took a New Year's resolution to take at least one roll of film, in whatever format, per month this year, but it is already April 1 (no foolin') and I haven't shot a single image on film. Soooo, I am playing with various B&W conversion routines. I half-thought that the solution to all this was a new Leica Monochrom, but after stewing about it for a couple of days, I couldn't see that the difference between the Monochrom and various B&W conversion algorithms was going to matter very much for this type of project. My final output is printed full frame in a 9x12 landscape Blurb book. The small size covers a multitude of resolution sins. . . frankly I don't know that you'd be able to tell the difference between a D100 file and an M9 file at that size.

So: no out of camera b&w. I have never liked the b&w that any digital camera produced, so it is back to PS5/Silver EFX for those files anyway.

 

[Benjamin Marks]

Here's a second go a the same scene. This time, light metered with a Luna Pro F, tripod f:5.6/1/4 sec. Composition is worse, exposure is better. Same conversion process, basically: NEF to DNG, Open DNG as 16 bit RGB in PS5 and convert to B&W in SFX. This time I left the image as an RBG file, but chose the default Pan-X profile. There are a couple of other subtle differences too. Overall, contrast is a little lower and "grain" is a little finer, I think because of what SFX is doing to the file.

I tried to correct a little of the AF-Nikkor's pincushion distortion in PS (somethnig that the Fuji does in-camera with its own lenses, I believe).