Author Topic: SP ABS Color Ring Scans (Read 174358 times)

Krogenar · « **on:** Sat, 13 July 2013, 10:22:48 »

My original project was inspired by Kaporkle's 3D renderings for groupbuys and thought it would be great if we could reliably reproduce the colors available from Signature Plastics. So I decided to try to determine the Pantone numbers for each chip. (http://en.wikipedia.org/wiki/Pantone) PMS (Pantone Matching System) is used by various industries to talk about color in clear terms. Mashby very generously bought me a SP ABS Color ring, and had it sent to me.

It was an excruciating process, trying to get Pantone numbers for each, and also very subjective. So I abandoned that project, and instead decided it would be easier (and much faster) to make high-res scans of each chip. My hope is that this might help people explore new color schemes. The files are large at 600dpi, and I think it's cool seeing these chips so large on a screen.

It had been pointed out by many people that these scans are not ever going to replace an actual color ring. Signature Plastics, GMK and other keycap manufacturer's always provide the caveat that the final colors are sometimes different from what's on the chip, and that the colors will change over time. Please keep this limitation in mind when using the scans.

UPDATED: (07/19/2013)

The files have been made available by Mashby on his Flickr photostream (thank you Mashby!) which can be found here:

http://www.flickr.com/photos/compactkey/sets/72157634628432952/with/9294353672/

The original high-res TIFF files can be found here, thanks to GH Good Guy Supreme, Mashby:

Blues: http://l.mashby.com/Zs44
Brown-Whites: http://l.mashby.com/cjGz
Colors: http://l.mashby.com/n3LJ
Greys 1: http://l.mashby.com/hydi
Greys 2: http://l.mashby.com/sdsm

UPDATED: (10/31/2013)

ijprest, with access to a ColorMunki colorimeter has performed some really great color scans! He has intergrated this color information in his Keyboard Layout Editor as selectable color palettes. Phenomenal work!

For now, I'm going to be cropping Dianoda's superior scans so that they look a bit more presentable.

UPDATED: (03/05/2014)

Thanks to Dianoda for providing scans of SP's PBT chips.

badcop · « **Reply #1 on:** Sat, 13 July 2013, 10:25:57 »

you probably looked into this but did you check to see if SP has RGB, Hexadecimal or CMYK for each color?

Krogenar · « **Reply #2 on:** Sat, 13 July 2013, 10:36:31 »

My original project was inspired by Kaporkle's 3D renderings for groupbuys and thought it would be great if we could reliably reproduce the colors available from Signature Plastics. So I decided to try to determine the Pantone numbers for each chip. (http://en.wikipedia.org/wiki/Pantone) PMS (Pantone Matching System) is used by various industries to talk about color in clear terms. Mashby very generously bought me a SP ABS Color ring, and had it sent to me.

It was an excruciating process, trying to get Pantone numbers for each, and also very subjective. So I abandoned that project, and instead decided it would be easier (and much faster) to make high-res scans of each chip. My hope is that this might help people explore new color schemes. The files are large at 600dpi, and I think it's cool seeing these chips so large on a screen.

I'll start with the 'greens' as they are the smallest group.

esoomenona · « **Reply #3 on:** Sat, 13 July 2013, 10:45:48 »

VAG is always number one.

mashby · « **Reply #4 on:** Sat, 13 July 2013, 11:40:58 »

This is awesome Krogenar. Thank you for putting so much time into this.

damorgue · « **Reply #5 on:** Sat, 13 July 2013, 11:45:15 »

Ideally, we should get the entire colour ring in one shot and with as many other scales present as possible. NCS, RAL, Pantone, CII and what have you so that people can always relate their colours to whatever colour scales they may have access to irl.

Krogenar · « **Reply #6 on:** Sat, 13 July 2013, 13:27:53 »

I wish there was some way to post a zip file -- I know I've seen someone else do it -- but the .zip files themselves range from 15 to 62 MB, so I'm not even sure if the forum software will allow it. I could do them individually, after converting from TIF files to JPG. Looks like that's going to be it for now. I'd like to upload the entire collection to the wiki at some point.
Here's another green tile.

Krogenar · « **Reply #7 on:** Sat, 13 July 2013, 13:28:27 »

And another.

Krogenar · « **Reply #8 on:** Sat, 13 July 2013, 13:29:09 »

(reaches into his lapel pocket)

Oh dizzam, look!

Krogenar · « **Reply #9 on:** Sat, 13 July 2013, 13:30:00 »

(Krogenar stares at the ABS chip, silently judging it.)

Photoelectric · « **Reply #10 on:** Sat, 13 July 2013, 13:36:24 »

Zip file with all the colors in it? Why not host them as jpg somewhere like Imgur (<--highly recommended) and then link them all in a few posts, one per color range? You can mass-select a lot of files to upload to Imgur in a row.

mashby · « **Reply #11 on:** Sat, 13 July 2013, 14:21:02 »

Doh! I can post these on Flickr too if you like. I never thought of that, but would be happy to at native resolution.

Techno Trousers · « **Reply #12 on:** Sat, 13 July 2013, 15:02:15 »

Somebody with access to a spectrophotometer needs to step up on this one.

mkawa · « **Reply #13 on:** Sat, 13 July 2013, 19:04:13 »

dianoda and i have been working on this krog. the problem with scanners is that they are quite difficult to color calibrate (although i don't know, maybe you have a drum or flatbed that _is_ calibrated with an appropriate measurement device?

however, recall that flatbed scanners are just upside down copy stands with a motor. that is, they're cameras. but, for some dumb reason, i have yet to find a flatbed scanner that will just give you its raw sensor output.

thankfully, actual cameras are much better about this, and at the very least. actual light source color temperature can be measured quite easily and corrected for. additionally, if you have a fully calibrated chain (camera, monitor) and high quality chips of known emissivity (eg, a pantone swatch), you can build a corrective profile to apply to that sensor output.

anyway, to this end, i sent dianoda a set of abs chips last week and he's going to start on them shortly, if he hasn't already. two sources of imaging would actually be nice for comparison purposes, and process refinement but krog, you starting on the PBT chips first would be more expedient in terms of getting _some_ relatively accurate values to work with. in particular, matt3o wants to pull averaged RGB data for his key picking front-end.

mashby · « **Reply #14 on:** Sat, 13 July 2013, 20:49:11 »

I took the liberty of putting together a Flickr album with all the individual scans. You can find it at this link. Fantastic work Krogenar. I really like how they came out!

They need to be reordered a little bit, the upload process took things a little bit out of order from the way Krogenar sent them to me. I moved them out of my DropBox, so I'll have to sort them when I'm back at the office on Monday.

OAV by cpkey, on Flickr

GC by cpkey, on Flickr

Photoelectric · « **Reply #15 on:** Sun, 14 July 2013, 09:28:01 »

Those are beautiful, Krogenar (just looked at mashby's hosted photoalbum). Great work, and is definitely very helpful!

Matt3o · « **Reply #16 on:** Sun, 14 July 2013, 09:34:33 »

This is a great project! It would benefit both the keyboard color chooser and the what the abs projects. Also would be great to have the same for PBT!

BUT...

PLEEAASE add a the same pure white reference for each chip! Also probably a scan of all chips together would be far more accurate in terms of post-calibration!

Krogenar · « **Reply #17 on:** Sun, 14 July 2013, 10:22:24 »

Quote from: mkawa on Sat, 13 July 2013, 19:04:13

dianoda and i have been working on this krog. the problem with scanners is that they are quite difficult to color calibrate (although i don't know, maybe you have a drum or flatbed that _is_ calibrated with an appropriate measurement device?

You're right -- color is very difficult to quantify at times. Someone mentioned a spectrophotometer, but man are they expensive. When I was trying to do PMS matching, I tried to create a standard viewing area to make it easier using one of those 'sunlight' producing lamps, and it was just mind-numbing.

Quote from: mkawa

however, recall that flatbed scanners are just upside down copy stands with a motor. that is, they're cameras. but, for some dumb reason, i have yet to find a flatbed scanner that will just give you its raw sensor output.

I acknowledge that these scans won't be perfect, but they're a start at least. The way I do it for my own business is this: I print out a gamut (about 300 colors swatches) on the substrate and change spot colors in the file if needed. These scans look color-accurate to me, but they may not be perfect. If your monitor is different from mine then they'll look different on your screen. But I don't see any way around that aside from using a spectrophotometer to take CMYK or Lab values.

Quote from: mkawa

thankfully, actual cameras are much better about this, and at the very least. actual light source color temperature can be measured quite easily and corrected for. additionally, if you have a fully calibrated chain (camera, monitor) and high quality chips of known emissivity (eg, a pantone swatch), you can build a corrective profile to apply to that sensor output.

Yeah, I noticed that a lot people just photograph the chips and post them with groupbuys. My use of them is really just to shuffle them around on a screen, so I'm not worried about them being perfect, just usable.

Quote from: mkawa

anyway, to this end, i sent dianoda a set of abs chips last week and he's going to start on them shortly, if he hasn't already. two sources of imaging would actually be nice for comparison purposes, and process refinement but krog, you starting on the PBT chips first would be more expedient in terms of getting _some_ relatively accurate values to work with. in particular, matt3o wants to pull averaged RGB data for his key picking front-end.

It took about 2 hours to do everything, and I have the original 100MB scans as well, before I cropped them all apart, in the event that we need to color correct the images, as a group. I'm willing to do a PBT ring as well.

Krogenar · « **Reply #18 on:** Sun, 14 July 2013, 10:28:04 »

Quote from: Matt3o on Sun, 14 July 2013, 09:34:33

This is a great project! It would benefit both the keyboard color chooser and the what the abs projects. Also would be great to have the same for PBT!

BUT...

PLEEAASE add a the same pure white reference for each chip! Also probably a scan of all chips together would be far more accurate in terms of post-calibration!

Here's what I did -- I have a CanoScan 5600 flatbed scanner. I placed the chips into groupings on the scanner bed, keeping them as square as I could for post cropping, then scanned them at 600 dpi. Saved them all as TIF files, and then started the process of cropping them out. The images I've posted in the thread are JPGs with the quality settings maxed out. The bottom of the scanner 'lid' has a white cushion, so that's what you would be seeing in the background of the larger scans. How would I add a pure white reference to this process to improve it? I have white posterboard card stock that I could use as a backing -- would that suffice?

Oh, and someone please tell me why I've fallen in love with this color:

It's not a pink, per se, nor is it really a red or an orange. It's a salmon color -- so far the most intriguing color on the ring to me.

Mashby, thanks so much for putting up the Flickr channel, that's awesome. Matt, I didn't know you had made the 'What the ABS' app, it's great! I think I may be able to get my hands on a spectrophotometer. What workflow is going to yield us the best results? Scanning each chip directly?

Matt3o · « **Reply #19 on:** Sun, 14 July 2013, 10:34:02 »

Quote from: Krogenar on Sun, 14 July 2013, 10:28:04

Here's what I did -- I have a CanoScan 5600 flatbed scanner. I placed the chips into groupings on the scanner bed, keeping them as square as I could for post cropping, then scanned them at 600 dpi. Saved them all as TIF files, and then started the process of cropping them out. The images I've posted in the thread are JPGs with the quality settings maxed out. The bottom of the scanner 'lid' has a white cushion, so that's what you would be seeing in the background of the larger scans. How would I add a pure white reference to this process to improve it? I have white posterboard card stock that I could use as a backing -- would that suffice?

basically what I'd need is at least all colors of the same kind (reds/greens/blues/...) scanned together with the same white reference. The whiter the better (matte, not glossy). If you can't find a good enough white reference at least place the whiter color chip you have on all scans.

I know this is pretty empirical but at least it will be better than the ****ty pictures we have so far.

THANKS!

mkawa · « **Reply #20 on:** Sun, 14 July 2013, 10:57:19 »

i have an i1 display 2 that has been sitting for a while. i sent my abs ring off to dianoda because i'm getting another one with a few more colors the second SP makes another batch. i will be able to do output comparisons of all the images we create.

mkawa · « **Reply #21 on:** Sun, 14 July 2013, 10:58:52 »

note that my display 2 is 4-5 years old and no longer nearly as accurate as it once was. i have it sealed in a bag with dessicant, but i'm not going to pretend that that's enough to preserve its accuracy. we'll just have to do the best we can *shrug*

Matt3o · « **Reply #22 on:** Sun, 14 July 2013, 11:05:32 »

Also colors may change over time. So at the same SP color code may not corresponds the same real/Pantone color

Krogenar · « **Reply #23 on:** Mon, 15 July 2013, 06:30:01 »

Quote from: mkawa on Sun, 14 July 2013, 10:57:19

i have an i1 display 2 that has been sitting for a while. i sent my abs ring off to dianoda because i'm getting another one with a few more colors the second SP makes another batch. i will be able to do output comparisons of all the images we create.

That is awesome. I have an XRite i1, but not the display 2 version. I think it might be able to give me a CMYK breakdown of each chip, and it does have a special 'white' base chip to scan, for calibration purposes.

Oh, and Matt, here's a low-res JPG of the one of my scans:

Again, this is a low-res JPG. The original file is a 600 dpi TIF, 49 MB, 2512x6864 pixels. I can supply you with these files Matt, if that would help. The subsequently cropped files are taken directly from these source files, with no processing or alterations, just cropping.

Matt3o · « **Reply #24 on:** Mon, 15 July 2013, 06:37:37 »

that is surely a better picture than what we have now. I'll try to work on that and see what I can do with it, it misses a white reference though, I don't think I can use the bottom right white as reference

Krogenar · « **Reply #25 on:** Mon, 15 July 2013, 06:51:49 »

Quote from: Matt3o on Mon, 15 July 2013, 06:37:37

that is surely a better picture than what we have now. I'll try to work on that and see what I can do with it, it misses a white reference though, I don't think I can use the bottom right white as reference

I'll send it (the original) along to mashby, and hopefully he can upload it to the Flickr channel for you.

EDIT: Successfully uploaded to mashby -- thanks again, mashby, for hosting these files. Once we get this all worked out, we should probably try to get them on the wiki.

mkawa · « **Reply #26 on:** Mon, 15 July 2013, 07:46:29 »

i've never tried using the display 2 on anything but a backlit display. i'm not sure that it's particularly accurate without a good 75-100lm (actually that's probably the wrong unit, but you get the idea) of light into the pigments.

iirc the display 2 (and similar RGB spectrophotometers) is 3 photosensors that are pigmented to a mean of R, G and B. each pigment energizes at the given mean wavelength. some measurement of energy (either differential or charge) is read to determine emitted light color. two major downsides to this kind of sensor: 1) limited lifetime. the pigments are not stable, and are particularly sensitive to humidity. 2) accuracy is very low without a heck of a lot of energy into the sensor.

there's also some issue with gamut. for example, it's apparently not really very accurate on my 98% aRGB u2711. i've been too lazy to calibrate for quite a while, so ymmv and mmmv. lol

mkawa · « **Reply #27 on:** Mon, 15 July 2013, 07:54:53 »

actually speaking of i1, they now offer a shoot and click camera profiling solution bundle with the new display pro (the upgraded display 2)

http://www.amazon.com/X-Rite-EODIS3CCPP-Display-ColorChecker-Passport/dp/B007V9N65O/ref=sr_1_1?ie=UTF8&qid=1373892732&sr=8-1&keywords=i1+display+pro

tempting and not much more expensive than the display 2 was.

Krogenar · « **Reply #28 on:** Mon, 15 July 2013, 08:04:35 »

Quote from: mkawa on Mon, 15 July 2013, 07:46:29

i've never tried using the display 2 on anything but a backlit display. i'm not sure that it's particularly accurate without a good 75-100lm (actually that's probably the wrong unit, but you get the idea) of light into the pigments.

iirc the display 2 (and similar RGB spectrophotometers) is 3 photosensors that are pigmented to a mean of R, G and B. each pigment energizes at the given mean wavelength. some measurement of energy (either differential or charge) is read to determine emitted light color. two major downsides to this kind of sensor: 1) limited lifetime. the pigments are not stable, and are particularly sensitive to humidity. 2) accuracy is very low without a heck of a lot of energy into the sensor.

there's also some issue with gamut. for example, it's apparently not really very accurate on my 98% aRGB u2711. i've been too lazy to calibrate for quite a while, so ymmv and mmmv. lol

Even if we can get RGB/CMYK values for the chips, it never looks quote right, going from an illuminated LCD screen to reality just never perfect. I just figure if it's easier to see these things to some degree of accuracy it will mean more interesting groupbuys. When I did Pantone matching for the CCoG groupbuy's kaporkle renderings, I think the colors were much brighter than they should have been. But so long as we provide that caveat, it should work.

I'd love to know where these color codes (the names) come from. Is "GGK" a way of saying, put in 2 parts black, and one part of something else? It must be. Maybe it's a 60/30/10 code? 90% of "G" pigment and 10% of "K" pigment? That's my guess. Maybe we can do scans with the whitest chip (a known value) in order to see how the scans should be corrected?

Matt3o · « **Reply #29 on:** Mon, 15 July 2013, 08:19:48 »

Quote from: Krogenar on Mon, 15 July 2013, 08:04:35

Maybe we can do scans with the whitest chip (a known value) in order to see how the scans should be corrected?

THIS!

mashby · « **Reply #30 on:** Mon, 15 July 2013, 11:00:23 »

I posted the .TIFF file to the gallery.

Master Scan 1 by cpkey, on Flickr

Krogenar · « **Reply #31 on:** Mon, 15 July 2013, 11:40:09 »

Quote from: mashby on Mon, 15 July 2013, 11:00:23

I posted the .TIFF file to the gallery.

Show Image

Master Scan 1 by cpkey, on Flickr

Thank you sir!

So, Matt, if I repeated these scans, but with each scan including the *ahem* "whitest" ABS chip (WCK? which one is it? Disturbing in light of my recent posts on the Trayvon Martin verdict lately, LOL) would that be sufficient? I have a few more of these sorts of scans -- would rescans be necessary, or could you determine the level of correction needed by checking the existing scan that already has the white chip in it? Let me know. I don't want to crush mashby's dropbox account completely. :)

EDIT: Mashby, you've got my scan of the white, black and brown group of tiles.

(Krog's phone rings)

"Hello, Rev. Sharpton." :mad:

Matt3o · « **Reply #32 on:** Mon, 15 July 2013, 12:07:12 »

keep scanning with a white chip reference, I'll try to apply the same calibration to the picture you already posted. I hope it is going to work (if the scanner doesn't add some weird color correction by itself...)

Krogenar · « **Reply #33 on:** Mon, 15 July 2013, 12:21:53 »

Quote from: Matt3o on Mon, 15 July 2013, 12:07:12

keep scanning with a white chip reference, I'll try to apply the same calibration to the picture you already posted. I hope it is going to work (if the scanner doesn't add some weird color correction by itself...)

If Mashby can post the original scan that includes the white chip, would that be enough? This is the white chip with a number of other chips at the same time, not the cropped version. I figure whatever color shift is present will be consistent across all scans? I just sent it to his account, and it's only 20 MB.

Matt3o · « **Reply #34 on:** Mon, 15 July 2013, 12:48:19 »

Quote from: Krogenar on Mon, 15 July 2013, 12:21:53

If Mashby can post the original scan that includes the white chip, would that be enough?

absolutely

Quote from: Krogenar on Mon, 15 July 2013, 12:21:53

I figure whatever color shift is present will be consistent across all scans? I just sent it to his account, and it's only 20 MB.

I have the feeling that they tend to add some kind of calibration based on the scanned colors, but I guess it can be disabled.

mashby · « **Reply #35 on:** Mon, 15 July 2013, 13:15:56 »

The browns and whites are posted.

Master Scan - Browns & Whites by cpkey, on Flickr

Krogenar · « **Reply #36 on:** Mon, 15 July 2013, 13:40:37 »

Mashby, matt, all that's left are the blue chips (all on one scan) and the grays (one large scan, and a smaller). I'm in the processing of uploading them to Mashby now.

EDIT: Remaining three scans (2 for gray chips and 1 for blue chips) are now uploaded to Mashby.

mkawa · « **Reply #37 on:** Mon, 15 July 2013, 15:59:42 »

Quote from: Krogenar on Mon, 15 July 2013, 08:04:35

I'd love to know where these color codes (the names) come from. Is "GGK" a way of saying, put in 2 parts black, and one part of something else? It must be. Maybe it's a 60/30/10 code? 90% of "G" pigment and 10% of "K" pigment? That's my guess. Maybe we can do scans with the whitest chip (a known value) in order to see how the scans should be corrected?

no, i talked to melissa about this at length when i was working on this: http://wiki.geekhack.org/index.php?title=Signature_Plastics_Parts_Numbers

the codes are somewhat arbitrary (not quite, since similar colors get similar codes) and code to a specific plastic purchased from a manufacturer of colored plastics. they don't mix these colors themselves at SP (it would be too hard to achieve uniformity given their tooling).

you, out of everyone, krog, should know that white is never quite white

. measurement is the only way we're going to get absolute accuracy (relative to our measurement devices HAH). that's why i like that we have multiple samples incoming right now; your scans, dianoda's photos, and so on. basically, take the literal mean of all these "reasonably well calibrated" images in the same color space, and you'll get the least common denominator of what your keys will probably look like

speaking of the wiki page, i think there's going to be some weird stuff happening wiki-wise shortly, but i also believe a migration path will be provided, so go ahead and add this content to that page (or add that content to this thread; either way is fine!

))

mashby · « **Reply #38 on:** Mon, 15 July 2013, 17:22:07 »

The remaining three master scans have been posted.

Master Scans - Greys #1 by cpkey, on Flickr

Master Scan - Greys #2 by cpkey, on Flickr

Master Scan - Blues by cpkey, on Flickr

Krogenar · « **Reply #39 on:** Mon, 15 July 2013, 20:34:09 »

Quote from: mkawa on Mon, 15 July 2013, 15:59:42

you, out of everyone, krog, should know that white is never quite white

Too true -- needs a speck of blue usually.

Quote from: mkawa

measurement is the only way we're going to get absolute accuracy (relative to our measurement devices HAH). that's why i like that we have multiple samples incoming right now; your scans, dianoda's photos, and so on. basically, take the literal mean of all these "reasonably well calibrated" images in the same color space, and you'll get the least common denominator of what your keys will probably look like

Agreed. Caveats all around! LOL

Quote from: mkawa

speaking of the wiki page, i think there's going to be some weird stuff happening wiki-wise shortly, but i also believe a migration path will be provided, so go ahead and add this content to that page (or add that content to this thread; either way is fine! ))

Migration path? Ok, girding myself for impact. A hint, perhaps? A new wiki software is coming?

mkawa · « **Reply #40 on:** Tue, 16 July 2013, 09:16:16 »

ask rknize. i said my piece on this, but i'm too busy to do the engineering needed. he is trying to figure out how to centralize our information a bit better, that's all. MW doesn't seem to be doing the job *shrug*

MOZ · « **Reply #41 on:** Tue, 16 July 2013, 13:13:16 »

Just wanted to say excellent work guys.

Matt3o · « **Reply #42 on:** Wed, 17 July 2013, 09:40:51 »

interestingly WFK seems the whitest white, I though it was WA.

anyway I don't seem to be able to calibrate those images in a satisfactory way, especially whites and grays.

mkawa · « **Reply #43 on:** Wed, 17 July 2013, 11:34:33 »

i think the way forward for right now (i am eyeing the i1 display pro damnit, but i'd need someone to buy my old display 2 off of me...) is to wait for dianoda's images, pull the RGB values off both images and take a weighted mean.

Krogenar · « **Reply #44 on:** Wed, 17 July 2013, 11:53:11 »

Quote from: Matt3o on Wed, 17 July 2013, 09:40:51

interestingly WFK seems the whitest white, I though it was WA.

WFK also seems the whitest on my monitor, with WA having a very faint magenta tint. I don't have the chips with me at the moment -- how do they compare to the ones you have Matt (assuming you have them with you)?

Quote from: Matt

anyway I don't seem to be able to calibrate those images in a satisfactory way, especially whites and grays.

How are you trying to calibrate?

I don't think you're working with the TIFFs. Let me see if I can share the files with you via dropbox -- would it help to have the full scans? I'm not entirely sure that flickr is displaying the TIFF, or downsizing it somehow.

EDIT: from the Flickr stream, you can access the really large images, but they're JPGs, not TIFF files. I'm not sure if that would matter to your calibration, Matt, but in either case, the WA and WFK appear the same.

Krogenar · « **Reply #45 on:** Wed, 17 July 2013, 12:02:05 »

I'm referencing this link from the Raindrop GB -- looking around for other photos of chips to see how mine stack up:

http://www.keycapsdirect.com/images/Customized/ABSColors.jpg

The ~~WCK~~ (gah!) WFK chip looks very, very white as well.

mashby · « **Reply #46 on:** Wed, 17 July 2013, 12:08:18 »

Quote from: Krogenar on Wed, 17 July 2013, 11:53:11

EDIT: from the Flickr stream, you can access the really large images, but they're JPGs, not TIFF files. I'm not sure if that would matter to your calibration, Matt, but in either case, the WA and WFK appear the same.

I didn't realize that Flickr would convert the images. I'll post the native files on my server for those that want to download them.

Edit: Feel free to put these URLs in the OP if you like.

Blues: http://l.mashby.com/Zs44
Brown-Whites: http://l.mashby.com/cjGz
Colors: http://l.mashby.com/n3LJ
Greys 1: http://l.mashby.com/hydi
Greys 2: http://l.mashby.com/sdsm

Krogenar · « **Reply #47 on:** Wed, 17 July 2013, 12:50:20 »

Quote from: mashby on Wed, 17 July 2013, 12:08:18

Quote from: Krogenar on Wed, 17 July 2013, 11:53:11
EDIT: from the Flickr stream, you can access the really large images, but they're JPGs, not TIFF files. I'm not sure if that would matter to your calibration, Matt, but in either case, the WA and WFK appear the same.

I didn't realize that Flickr would convert the images. I'll post the native files on my server for those that want to download them.

We don't even know if that's an issue, though, Mashby. When I open the JPG and the TIFF file alongside each other, they look identical. But again, I'm not sure how Matt's trying to calibrate. I did look up some more GBs, and found that Raindrop and The Valentine sets both used the ~~WCK~~ (Gah! AGAIN!) WFK chip as the white chip, which would only seem to make sense if it was the whitest chip, right?

WFK = "What the F*** Krog!" LOL

mashby · « **Reply #48 on:** Wed, 17 July 2013, 12:53:36 »

Better safe than sorry. I posted direct links to the .tiff files in my earlier post.

Matt3o · « **Reply #49 on:** Wed, 17 July 2013, 13:15:17 »

Quote from: Krogenar on Wed, 17 July 2013, 11:53:11

WFK also seems the whitest on my monitor, with WA having a very faint magenta tint. I don't have the chips with me at the moment -- how do they compare to the ones you have Matt (assuming you have them with you)?

I have a very limited number of chips around. Unfortunately I don't have WA.

Quote from: Krogenar on Wed, 17 July 2013, 11:53:11

How are you trying to calibrate?

Not easy since I don't have a white reference. I tried to use WFK as white reference, but it doesn't work. Basically it's a trial and error, comparing the colors to the ones I have. The scanner seems to have added quite a bit of "sepia" and colors seem a bit desaturated, but it is also true that color chips varies over time so the ones I have might not be the same exact colors that you have (even though they have the same code).

I'll make some other tests tomorrow and post the result

News:

Author Topic: SP ABS Color Ring Scans (Read 174358 times)