I work in a place with several Ptek DI presses. At the time we bought them, it made sense, as making that number of plates would have constituted a violet platesetter, and we still would have needed to buy the presses. We've had very good luck with the QMDIs, especially with that generation of lasers (circa 2001). I don't think we've had more than maybe one laser fail over four machines.
We're an inhouse mailing operation with a huge mailing responsibility with runs that are typically beyond what the DI plates can handle, so we also have a need for metal plates. When that operation went online at the same time, the CTP options were nowhere near as robust as they are now. It's enough of a difference that we needed to replace our existing Presstek CTP option with a newer, faster and fully automated machine.
What we see now with the new fully auto thermal CTP device is that there is little to no reason to have those DIs. The platesetter is fast enough that our narrow web press + our new platesetter is faster than the DIs to get off say, 1000 copies.
While I realize we're unusual in how we work, it's been very clear to us that the DI's have a very limited scope in the market. I suspect if you consider an automated plate hanging press, and even an *unautomated* fast thermal platesetter, you'd be within spitting distance of the DIs.
And you'd have more latitude in your choice of plates and setter(not just Presstek), and most importantly in your workflow. Presstek takes a standard 1bit tiff (at least in my scenarios) and rotates it 90 degrees (I think they figured that they'd only be dealing with Quark's non-PostScript compliant output methods) and adds a Presstek header label. Big deal. Thanks a lot for turning a generally compliant data format into something vaguely proprietary. What this means is that this press has RIP options on par with a color copier. And if you want to get around that, prepare to jump through arcane prepress hoops to do so.
The market has shifted substantially since the DI took off, and I think it's clear to those that own them, that the money would have been better spent keeping the imaging offline. Especially if you're going to output to more than just that DI.
I think a lot of the appeal for these presses comes from a press operator/owner mindset; that this bundled operation can nearly eliminate prepress by adding the platesetting to the press, etc. Now all that prepress has to worry about is imposition, right? Yeah, uhm, no. Files are still going to be awful to deal with, and now the prep operators don't have a say in the workflow they have to live and die with. And I'd almost guarantee that the plates offered will be significantly more expensive than any other alternatives.
If you're really considering a DI, you really have to look at what alternatives are out there and do an in-depth financial comparison, and not just at purchasing time, but down the road. Do you want to depreciate a CTP machine and press out separately and replace them at five and ten years respectively, or squeeze out ten years of a prepress/plate operation because you need to keep making money off that press? And just what is that depreciation going to look like compared to the conventional CTP&press arrangement?
It's a lot bigger question than just plate to plate register and on press imaging benefits.
