You shouldn't normally get any colour cast from a correct colour conversion. There are several "intents" for conversion from one space to another. Generally I use relative colormetric with black point compression.

In fact, in Photoshop you shouldn't even see any difference on screen when you convert between colour spaces.

The only possibilities are:

1) They assume sRGB if there is no profile attached in the JPG. In this case, make sure you save with profile attached (I reckon this is most probable)

2) Their profile is a bad one. If the lab is worth their salt, the profile should be up to date.

3) Your monitor calibration is out.

You can rule out 3 if the sRGB print looks the same as on screen sRGB in PS. As I said, it's probably option 1, and you can test this way:
* Convert to their profile in the normal way
* The assign to the sRGB profile (not convert)

If this shows the red-cast, then their equipment is expecting to see the profile

Hope that helps (a bit!)
Matt

If you shoot RAW, then it doesn't really matter what colour space your camera is set to.

The colourspace is used for JPG (and I believe the embedded JPG in the RAW too), so if you use a portable storage device (like the FlashTrax) then you're probably best leaving the colourspace in camera to sRGB as only the newest Epson device supports AdobeRGB.

FYI, AdobeRGB has a wider colour gamut compared to sRGB, and ProPhoto is even wider still (massively wide in fact which is why you need 16-bit per pixel, not 8 otherwise you risk posterisation). I'd advise using AdobeRGB if:

1. You intend to print (especially with Litho printing), and
2. You can get your head around colour management

(If you can't cope with profile conversion and selection, then just stick to sRGB )

Hope that helps!
Matt

Best is to convert to AdobeRGB (or 16-bit ProPhoto) from the RAW. Do your post-processing in PS and save.
You will find that their profile will (should) change over time and keeping your processed files in a standard profile will help - especially if you need to pass the images onto anyone else.

Then, just before saving the JPG for printing, use "Convert to Profile" in PS and use their profile. If you have CS/CS2 you can soft-proof to see what the effect will be before conversion (and check if there's anything out-of-gamut).
Matt

With digital you may be better off creating a "stack" by combining many reasonable exposures together. More work in PS, but the noise is easier to control that way. (Do a Google for astronomical stack and you should find a few ideas)
Matt

Will,
I've only used the full Developer Studio of Silkypix (should be "SILKYPIX"... not sure if the free version restricts TIFF. In the SP-DS3 you choose the file type (and bit depth) in the "Settings" when you convert.

* 16-bit vs 8-bit TIFF - if your images are great straight out of SP-DS3, then don't worry too much - use 8-bit. If you intend to do dodging and burning, curves etc in Photoshop CS2, then for the K10D I would advise using 16-bit TIFF. The K10D has amazingly clean shadows which you may want to fiddle with in PS - if you do this in 8bit then you risk posterization. Stick to 16-bit and you'll get smoother results for sure

* Colour. Most RAW convertors assume linear response from the CCD and really don't do much when converting. Capture One can apply a CCD profile before all the RAW conversion which produces better results. Silkypix goes a little further (or in a slightly different direction) by allowing the choice of colour responses - my "usuals" are "Memory Color 2", "Film P" and "Film A". You'll struggle to get the same colour response out of ACR

* Sharpening: I find that the *ist-D did very well using their "Pure Detail" default, and then adjusting the noise reduction to around 6 to 20 (depending on ISO). The *ist-D was getting better results with this than I'd ever seen from 6Mp. The K10D doesn't need as much, and the "Natural fine" (the default) seems just fine - but the noise reduction still needs to be around 4 to 8 to smooth out the sky areas. I'll still do some localised sharpening in PS though

I've been getting astonishingly good results from the K10D and SP-DS3 - you can see the quality at A4 (I've had MF level of detail), they'd be exceptional at A3 and beautiful even larger.

One day I'll get round to a "Rough Guide to Silkypix" like I did for Capture One....
(It might help if you keep badgering me with questions )

Matt

Hi all

Good news
http://www.isl.co.jp/SILKYPIX/english/download/early_preview/

ISL have released v3.0.3 preview that includes support for the K10D

Shouldn't be long for a full blown 3.0.3 release!

Matt

Just to add a little info:

The pop-up flash "fizzes" like that to provide an auto-focus assist - it's outputting bursts of light to enable the AF electronics to actually see what's going on. It's perfectly normal, if a little irritating

The AF360FGZ and AF540FGZ have a separate AF-assist lamp that shines a red-illuminated pattern onto the subject to assist focus. This is silent and much less intrusive!

And as has been said, the AF360/540 really are great pieces of kit, and the only real choice for the *ist-DL/DL2/K100D/K110D/K10D that lack TTL support (they only have P-TTL support).

Matt

Just checked the Metz site:
http://www.metz.de/downloadFile.php?type=adapterarchiv&lang=en
(It's a PDF)

And you'll see that:
* ist-D and *ist-DS can do TTL, A, M (note 9)
* All other DSLRs do A and M only (note 7)

I think the SCA 3702 will still pass the ISO rating and zoom info to the flash, but you're restricted as to the modes you can use. (I have an SCA 3702 going and a Metz 50MZ5 full system if anyone is interested )

Matt

AFAIK, the Metz will do "A" and "M" mode on the *ist-DL2.
With the SCA adapter you could do "TTL" on the *ist-D and *ist-DS (which was the last DSLR to support TTL flash).

The P-TTL mode is not supported by Metz on the Pentax at all.

Matt