inautilus wrote: ↑Sun Mar 08, 2020 5:40 pm
... it may be that the project is already locked into and is now committed to some other particular printing process (I don't know).
First, I comment on the
process:
The process started is a very informal test. And as you saw in the same weekend it was handed to the printer we changed 4 or 5 things again. That is part of this process. It is and must be iterative. So the only process in progress is the making of a template which involves laser cutting of the material, and printing on it as a test, to see if the cut lines are correct, if the concept works, etc. And if the printing is bright enough. The second time printing, I will use the same or different material, and I will no doubt have comments on changes on the graphic.
Nothing is locked into and we must navigate a way forward to suite ourselves.
The first process I see: The way I see it is that there are laser cutting of soft material done in every city almost. And printing. Once I can get a combination of material, machine, source files, and capable printer (person), I can document those steps and make that available. My cut line file will be free for use under GPL. If your template is covered under the same GPL, then there is a solution where users can make a template themselves or order one from another party who could make and sell them under the rules of GPL.
The second process I see is a follow up meeting of the meeting Dani had with Michael of SwissMicros in August 2019, for Swissmicros to make templates and sell new WP43C units. I also want to explore the possibility of SM selling a conversion kit, to convert a
DM42 to a WP43C. The steel plate which has the apertures for the buttons and the painted text is replaceable without opening the DM42, so it is possible. H2X had to replace his due to lifting of the edges with a new one from SM.
Third processes could well be further companies who can make templates, you have mentioned a very competent company for that.
inautilus wrote: ↑Sun Mar 08, 2020 5:40 pm
Several times now you have expressed concern Re potential issues when printing in fine detail. Are there going to be issues ... what is possible, and what isn't. Please forgive me if you've already 'been there, done that'.
That is all they are: potential. There are no hard rules. The best answer is it depends. And if you have the luxury of money, make more than one realistic test on the target medium. Then you know if the process works or not. There are many processes for marking equipment and protecting it. The ones I was involved in taught me the basic rule that if the lines are wider, more paint gets stuck and the less your chance for blotches is.
However, my first line of comment on layout is always technical printing related, i.e. the bigger the better.
My second line is always that it must be balanced. Spacing is not mathematical. It needs an 'eye' and involves juggling until the eye sees it right. Like a good photo: If you use thirds, it is sterile and if you balance it, it becomes alive. Relative sizes of letters and exponents is a balancing act.
inautilus wrote: ↑Sun Mar 08, 2020 5:40 pm
So, after only a quick scan, it seems that this is
not Mission Impossible after all. As an alternative to the double-shot key production for instance, HP switched over to printing ... and has continued executing high quality, durable, and exceptionally fine detailed printing on calculator keys (and faceplates) ... ever since. So, if they can do it ... then it can be done! Just pull out your humble 35s and check it out.
The question now is ... what is/was their production process.
I actually went to fetch my 35S and I agree, their lines are fairly thin. I would be happy with that kind of process. And the important thing is: If we ensured that the letters are legible and properly done, then, if the printing result exceeds your expectations, you would have an exceptionally legible and clear faceplate.
If you look carefully on the photo though, there are lessons from HP's HP35S layout. Personally I think they could have spent more time critically analysing what they have on there, and they can no doubt improve f[MODE], f[sqrt], f[2], and if you look hard maybe some more:
.
- IMG_3572a.png (161.3 KiB) Viewed 5274 times
HP35S
.
Lessons:
- Their caps letters are as large as it can be. Larger=better vision. In fact they are marginally larger than ours.The slanted surface lettering is very small in places and those printed ok.
- Their the-root-of-y letters are too small, too close together. I mean you only have a point to increase, but the eye notices that.
- Their root-of-y x and y are the same size. If you make the exponent even smaller, it is simply too small to be seen, leave alone the other reasons.
- Their x's are not controlled to be the same sizes for the same kind of functions. See f[RCL], f[MODE]. Also see their smaller x not consistent, see f[sqrtx] and f[1/x].
- Their white lettering on the buttons are VERY thick strokes, which is what is needed to get the paint to stick. My DM42 key strokes and square root line are SO thin and blotchy and I think that is really not up to scratch for the great machine it is.
- f[2] shows they used multiple sizes for the same class of functions, compare with f[6].This they did because their choice of font for lower case letters had suboptimal stems on the b, d, g, l, etc. They should have picked a font with shorter stems and they should have had the difference between caps and lower less pronounced.
- g[5] shows KM instead of km. I would rather die. Eeek.
- Roll up and down arrow heads are unnecessarily small. They had space. They should have looked.
All in all, they should have had a another internal review prior to committing to the master positives...
inautilus wrote: ↑Sun Mar 08, 2020 5:40 pm
"HP 35s Construction
(
https://en.wikipedia.org/wiki/HP_35s)
The case features many design elements from 1970s HP calculators such as the ground-breaking HP-65, including a black case with silver-striped curved sides, slope-fronted keys, and gold and blue shift keys. The faceplate is metal, bonded to the plastic case. The key legends are printed, rather than the double-shot moulding used in the vintage models.
It was initially supplied with a hard zippered clamshell case with a pocket for notes, and a printed manual,[8] but this was later changed to a slipcase made of vinyl-covered cardboard with elastic sides and velvet lining, and a CD-ROM containing a PDF manual.
I have the hard zippered case but time was not kind to the case.
Either way, a nicely built modern calculator with a depressing way of working. I literally never used mine for more than a few calculations.
Edit: fixed obvious typographical errors.
Edit2: thanks for pointing out the error, corrected
in red above.