VOGONS


First post, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t

This idea just popped into my head about 3 minutes ago, so bear with me.

I have many many old cards that came from scrap lots over the past 7-8 years and were sadly already stripped of their back plates. A relative of mine recently got a resin 3D printer of some kind and it dawned on me that it would probably be the ideal device for making blank back plates since it could probably make a whole "sheet" of them relatively quickly and they'd be far flatter than one made with a normal 3D printer. The main reason for doing this rather than just use blank metal slot covers is the simplicity and ease of cutting and drilling all the proper holes\cutouts. I am not a metal worker and I have very little experience (or tools) for doing this with steel. Plus I have a bad habit of lacerating myself any time I try to work with metal.

The measuring and laying out of the cutouts is also a bit challenging since everything has to be pretty accurate and straight for it to not look weird. If one cutout is in the wrong place then the card may not even slot in all the way.

Anyway... I got to thinking... Is there any combination of currently available websites, software or "AI" services that could be used to generate replacement back plates for old cards based on photos? Like, if it starts with a basic model like this and then I give it a picture of the back of a card to make holes. If the software was customized it could even let you manually add labels on the ports if your printer is good enough and the font isn't too small. The fact that nearly all single-slot PC cards would use an identical base model for the plate is what makes this seem feasible. It would give the program something to start with that doesn't change.

Thinking on it now, it seems ridiculous to expect a piece of software to not only understand what I want, but to also make accurate measurements... and yet, there is a lot of crazy stuff happening in this field these days. If it isn't possible now, maybe it will be soon?

Now for some blitting from the back buffer.

Reply 1 of 2, by paradigital

User metadata
Rank Oldbie
Rank
Oldbie

I really dislike the fact that as humanity we are now jumping on the “Use AI for everything” bandwagon for even the simplest of tasks.

3D modelling an I/O bracket is about the simplest 3D modelling you can do, as an amateur 3D printer owner back in 2020, the first things I printed were a motherboard I/O shield and an expansion card I/O plate (for a CF card holder), they must have taken no more than an hour each to model, and that was when I didn’t really have any experience!

I also wouldn’t be surprised if half of the common cards you can think of are already on thingyverse ore similar.

Reply 2 of 2, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t

Don't get me wrong, I'm not an AI fanatic. I don't currently use anything labeled "Ai" for anything at all personally, and I know that it's the current buzzword everyone is using... Mostly incorrectly.

Still, a ton of processes are being automated because people and businesses are dumping a lot of money and time into this field these days.

And I'm not talking about common cards. I'm talking about several dozen obscure cards just in my collection. I would imagine that it isn't a stretch to say that a handful of Vogons collectors like myself may have hundreds of cards with no back plates. I would guess that there are potentially tens of thousands in the hands of all PC hardware collectors.

I'm sure once you have the knowledge of how to do the modeling it wouldn't be terribly difficult to make a selection of basic cutouts (gameports, audio jacks, VGA ports or other less common ones) and design each bracket individuallyn using those, but it's one of those things that when done to hundreds of cards is an awful lot of work and time spent by the community undoing the damage caused by careless scrappers. And that isn't taking into account time spent repairing the cards themselves, which many of us already do.

I agree completely that it's probably good to know how to do this, and I will likely take a stab at it myself now that I know someone who can actually print such a model... But it would be really amazing if it could be automated, so that everyone could benefit from it. I don't believe any other aspect of repairing and restoring vintage hardware has the possibility of being simplified or automated by software in the way that this potentially could.

Now for some blitting from the back buffer.