@jbzfn@euronews "So, we'll open this one up and sure enough here, used disks that we will reformat, put a fresh label on, and then hopefully resell," he said."
Isn't there a fundamental issue on doing that in that the media itself degrades somewhat over time, exposure, etc etc?
@nazokiyoubinbou@jbzfn@euronews For 3.5" with their sealed metal flap and rigid case it shouldn't be too bad. Obviously reused media is gonna be more likely to be physically damaged, but at this point beggars can't be choosers.
Honestly, even the new old stock you find piecemeal right now is pretty terrible. I often have better luck with old floppies that have been sitting on a cupboard for 30 years than I do with crappy, late-day sealed stuff you can still buy on Amazon.
@MudMan@jbzfn@euronews My first thought was the metal flap would help protect it some, but don't forget the spindle area (or whatever it's called) is relatively open out of necessity (it has to be able to turn without the media itself having resistance against the plastic.) The flap doesn't exactly seal super tightly anyway.
And yeah, I get that beggars can't be choosers, but what I'd rather see is some small company working to make new disks from scratch.
@MudMan@jbzfn@euronews Actually, I was wondering if 5.25" and 8" disks might -- ironically enough -- perhaps be better protected. Yes they're relatively open, but usually they're stored in that sleeve of a very dry and absorbing paper material. I would imagine this helps wick moisture away actually. Even the spindle part goes into the sleeve. However, the sleeve isn't usually super tight, so I'm not sure there.
@nazokiyoubinbou@jbzfn@euronews I expect even a good floppy to have other issueslong before the spindle becomes an issue. If you're worrying about rust or ambient exposure, there are worse things that can happen to a floppy than moisture. There's nothing electric in there in the first place. Even back in the day I was able to extract data from muddy, damp floppies with rusty spindles or flaps. The magnetic disk itself will give out long before the physical media does unless very poorly stored.
@nazokiyoubinbou@jbzfn@euronews Here's a thing about old new tech: it tends to be either super expensive, terrible or impossible to reproduce.
Floppies aren't necessarily trivial to make or have enough demand to justify making them. There are new floppies out there, but I swear half of the ones in each box I buy are faulty out of the gate. Ditto for newer floppy drives, too. I've had much more luck restoring old drives than trying to use modern reproductions.
@MudMan@jbzfn@euronews Don't misunderstand me. I don't disagree with this. And it's good that they're not just having these straight up go to waste. I just worry that they can't necessarily be trusted that much more anyway. It doesn't sound like they have any specialized tests or methods of refreshing the media or anything like that, so for all you know when you get one it may have sat in a damp basement for 30 years.
@nazokiyoubinbou@jbzfn@euronews
Right, what I'm saying is that unless you're storing them in a swamp you won't have issues with mold or rust on the average floppy. At least not before the actual data storage gives up the ghost.
My source is living pretty much on a swamp and routinely using both 40 year old floppies and new floppies.
@nazokiyoubinbou@jbzfn@euronews Given that the disk itself is very thin, floppy plastic I expect temperature changes and warping would be a bigger threat than moisture in this particular case, and a plastic wrapper doesn't help with that.
So yeah, I'm guessing this guy testing and refurbishing old floppies (best guess is he degausses and rewrites the sectors raw using some of the modern tools to do that) actually increases the reliability of his stock quite a bit.
@nazokiyoubinbou@jbzfn@euronews FWIW, a ten pack of floppies on Amazon today is 20 bucks, and in my experience they tend to not work unless you do pretty much that same process on your end.
It makes perfect sense that a specialty seller working with airlines and manufacturing for legacy support would work this way. You can see similar stuff when it comes to FPGA or single board part replacements for those clients.
Hobbyist preservationists love those guys, they're keeping history alive.
@darth@MudMan@jbzfn@euronews I think the issue is a lot of them are very cheaply made in generic Chinese factories by those companies that pretty much disappear overnight one day.
@darth@MudMan@jbzfn@euronews In all fairness, as a bit of a tackon to this whole business, there are apparently floppy drive emulation devices that I guess use memory cards or whatever to mount images (don't know the specifics, but I've seen them around for Amigas and such in particular and at least one that I think was on an Apple.) As this can breath new life into such devices more reliably it might also affect demand. Of course it's not the same as holding and using a real disk...
@MudMan@darth@jbzfn@euronews One thing I'd like to see emulators do (and I've even suggested it directly on Dosbox-X's Github) is to emulate the sounds of a harddrive and floppy drive -- especially the harddrive. I admit I do very much love the idea of powering on an emulator and hearing that harddrive spinup sound they used to make as they came on.
I would hope if one does it, eventually others will as well.
@MudMan@darth@jbzfn@euronews Ah. If I'm to be honest I was a PC kid, not an Amiga one, so I never particularly got into Amiga stuff. In particular I really like having more powerful MIDI options.
@darth@MudMan@jbzfn@euronews Well, they weren't really trash at any point, but it was trickier. Amiga gave you a lot built in, but PCs ultimately offered more options and power.
@nazokiyoubinbou@darth@jbzfn It depends on where you were, when you were born and how much disposable income you had.
Amigas here were expensive, we were in 4-5 year old PC hardware at BEST. I never got a sound card until the Pentium generation, it seemed like an extravagant expense.
Old computing moved very fast and a lot of stuff was deplying unevenly or simultaneously in many places. The modern timeline many people quote is just not universal at all.
@darth@MudMan@jbzfn@euronews Do you mean a PC speaker? That was on most if not all "IBM Compatible" systems. I had a 286 with one for example. If you mean soundcards, like I said, that's one of the differences between Amiga and PC. With PCs you had options, but they came externally.
@darth@MudMan@jbzfn@euronews Basically the answer is no. The PC speaker was mostly just an indicator. Games did usually support using it to produce very simplistic sound effects, but for any decent sound you bought a soundcard.
That's what I mean about the differences. PC gave you more options, but you had to buy them. Many got very expensive. Some were worth it if you could pull it off.
But again, my timeline is skewed, because at the time if I had splurged for an AdLib I would have easily been able to slot it into my 286. Those generations of CPUs were in use a lot longer than people realize. 286s or 386s didn't go away the moment the replacement came out, like they do these days.
@darth@nazokiyoubinbou@jbzfn It was such a messy time. You had people stuck on Spectrums or CGA PCs deep into the 90s while others were on late-day Amigas or beefed up 486 PCs.
On home computing, at least, I don't know that it makes too much sense to think of side-by-side generational comparisons until what? 3D accelerators enter the picture?
@darth@MudMan@jbzfn I think the main issue for them might have been fairly stiff competition and starting to fall behind in what they excelled at. They were initially billed as a multimedia system (largely because of stuff like those built in graphics and audio capabilities) but as PCs started being able to drive VGA extremely well and even do SVGA (with 8-bpp or more) in some things as well as powerful soundcards, Amigas really kind of weren't ahead on that anymore.
@darth@MudMan@jbzfn@euronews You might be surprised. My first PC was a 286 (a blazing fast 16MHz!) and the soundcard I was given for it was an AdLib (can't even do digital sound!) Very limited and weak, yet definitely still capable of far more than you may realize. I found a surprising number of very fun games nonetheless.
@darth@MudMan@jbzfn@euronews There isn't a whole lot to innovate in quite such extremely obvious ways. However, there have been things going on a little less obviously. For example, AMD's switch to a SoC controlled chip was a wonderful idea that opened up enormous possibilities for performance handling (I'm still confused why Intel won't do this.) Most of the innovation goes to videocards these days and you already know about most of that.
Yeah, Imagine the dread of making a mistake and having the entire industry blaze past you on new tech, as opposed to having the competition get a 10% performance bump.
Intel just rebadged last generation with a slight overclock and they're not even gonna go out of business.
Making tech in the 90s must have been so confusing and stressful. Just zero safety net and nobody understands why things happen.
@MudMan@darth@jbzfn "Intel just rebadged last generation with a slight overclock and they're not even gonna go out of business."
What saddens me is that Raptor Lake is, in itself kind of a mistake (screws over some legacy compatibility) without a SoC design to handle instructions (making the OS do it) but I'm a bit bugged that both AMD and Intel are focused solely on benchmark wars right now rather than true innovation.
@darth@MudMan@jbzfn APUs also drive the latest consoles. Honestly though, it's not that much better versus separate chips for GPUs. Mostly it just helps concentrate things more simply.
Honestly, the real innovation of APUs has been thrown away and ignored. Specifically such chips could have been used to provide multi-architecture computing in one package with the best of both worlds -- heterogeneous computing. I guess things fizzled.
@darth@MudMan@jbzfn True, but imagine the real potential here. Imagine a full desktop with an APU processor AND a dedicated GPU. Normally the GPU portion of the APU is ignored, but imagine if operations that perform better on GPU cores were automatically routed to that and operations that performed better on CPU cores were routed to that and meanwhile the dedicated videocard was freed up to do its own thing with less load from stuff like physics.
@darth@MudMan@jbzfn As far as I can tell not really. To some extent if you have a similar enough GPU the system can basically not actually disable the GPU part of the APU, but as far as I know nothing truly utilizes it. PhysX does have a setting to run entirely on the CPU if you want, but it's not going to use the APU's GPU portion if you set that.
As far as I know nothing uses the potential and the GPU is basically disabled if you have a separate videocard for all intents and purposes.
@darth@MudMan@jbzfn Probably. But I don't think anything does it. Most don't even try to target and just use whatever is primary. Few things offer enough control that you could say "do this on the APU but that on the GPU."
I swear I read that AMD was going to make a compiler for heterogeneous computing that would just utilize everything fully, but I don't think it ever happened (or it must have not been very useful if it did I guess.)
@darth@MudMan@jbzfn Some things let you choose which thing you run them on entirely, but that's the entire thing. I'm saying individual components of one thing should run separately. For example, say, put physics on an APU's GPU portion and use the GPU entirely for graphics. Any settings you normally get are all or nothing. Put it all on one or put it all on the other. No mixture.
@darth@MudMan@jbzfn Nah, I can't agree with that. APUs are basically the same basic cores as GPUs so don't really add significantly to targets since basically it's the same as targeting a normal AMD dedicated GPU.
It really is just a matter of routing certain things certain ways. It can be done, just no one seems to be putting in any real effort to do it. APUs have largely disappeared as a meaningful thing from the desktop market, so there is no real incentive to now anyway.
You'll note it does have weaker stats. The max boost is 4.4GHz and its TDP design is intended for a range (most things are likely going to target the 45 rather than the 65.)
That's also not a lot of graphics cores really.
Also, I mentioned them competing over benchmarks. The Zen 3 design pushes heat so much that the GPU portion will likely drive the CPU down a lot (and vice versa.)
Though, all this said, as far as I know nothing can truly utilize this. I imagine this is, more than anything else, only a way to make sure that they all have minimal functionality if a dedicated GPU fails.
@darth@MudMan@jbzfn Yeah, I'm fairly certain the goal here is just to have something that can handle very very basic functionality if one does not have a dedicated GPU. It may beat out the really basic Intel onboard devices for example, but won't compete with normal onboard AMD or nVidia GPUs.
My bet is their processes have gotten to a point where it's cheap enough and easy enough that they just shrugged and said why not.
@darth@MudMan@jbzfn Exactly. Very basic 2D functions will work fine enough. You can probably even basically watch videos on it.
I don't really 100% know what they mean when they say "graphics cores" there. Actual GPUs list out separate types of cores. But, for example, the lowly Radeon 6400 has 12 compute and ray units. (I assume we're not talking about stream processors which should be in the hundreds.) I didn't feel like digging up laptop processors, but even that 7 is pretty low end.
@MudMan@darth@jbzfn Yeah, that's true. Heck, I have a "GPDWin" device that is a handheld built on the very short lived Intel Atom SoCs (actually I'm still really mad about them abandoning them as they were actually incredible for a lot of devices and a wonderful bang for the buck option making cheap devices viable even then. I have a 2-in-1 running on one that only cost $200 I still use today.)
Steam did greatly improve making them viable as a real platform though.
@nazokiyoubinbou@darth@jbzfn Sure, Valve lent its significant PR muscle to promote PC handhelds and its deep pockets to sell one at a loss to a mass audience. They were neither the first nor the best, though. Credit where it's due.
@darth@MudMan@jbzfn I'm a little unclear. You say "their APU" design, but AMD made the APU... As for OS, Valve took a fairly standard Linux setup and customized it a bit, which is hardly new (SONY has been doing this on their consoles since forever for example. Even the PS2 has a hint of Linux in its APIs and such.) Even the formfactor predates it.
Valve is the one that put it all together and sold it at a price people could afford while pushing for better gaming on Linux.
@darth@MudMan@jbzfn I don't see why any of them should get credit for those APUs, no. They didn't design them or anything. They just requested a specific basic design principle (probably specifically TDP, power usage, etc) and had the money to get it produced. They didn't actually DESIGN the processors, just select a few basic specs. AMD built them. And yes, AMD gets the credit for the Xbox and Playstation APUs. Similarly, you can't credit SONY for the Cell, just for its usage.
@darth@MudMan@jbzfn Right, this is fair. Valve was the one that took the idea (probably borrowing more from the Switch's popularity than Ayaneo/etc would be my bet, but as stated it is predated) and made it actually viable. They had the capital, resources, and market to pull it off and they did it and did it well. They didn't lock it down or anything like that, so it remains an actually good choice for its range.
@darth@MudMan@jbzfn Yeah, this is actually bigger than many even realize. By keeping it open they've made it a more overall useful platform and more available to devs and etc. Most with their size and capital choose to make locked down devices CoughMicrosoftAppleEtcCough but they didn't.
Honestly I give them credit for that above all else. Not just for being open on its own, but for being in the position to make it locked and get away with it yet not doing so.
@darth@MudMan@jbzfn Which is to say, all those others were unlocked devices of course -- but they had to be. They were small companies/kickstarters/etc. They didn't have the market to be able to push a locked down platform. Valve could have done it. Normally such companies never fail to succumb to such an urge.
@nazokiyoubinbou@darth@jbzfn Meeeeh, to a point. There were many lofty claims of how much support third party solutions, from Windows to SteamOS support on other Linux PCs, was going to exist that have been obviously missed.
The community has picked up some of the slack because of the open core of the project, but that doesn't mean that the endlessly open Windows competitor for gaming Valve seemed to be pitching isn't taking a back seat and breaking promises.
@MudMan@darth@jbzfn I do want to say though, Microsoft is making Windows so terrible these days that I really desperately hold out hopes that things like Proton and etc can make Linux really happen for gaming. It's not there yet and maybe it won't, but I just can't emphasize how badly we all need it. Especially now that MS has gotten to the point they lock make features exclusive to continuing to buy their latest OS (while having anti-consumer measures getting worse and worse in each.)
@darth@MudMan@jbzfn You can't credit Manjaro developers for making Linux. You can credit them for putting together the specific package and making it workable in its particular form and goals. And similarly, that's what Valve has done and what they deserve credit for.
@darth@nazokiyoubinbou@jbzfn
I'm not sure it's a good comparison. The SteamDeck is far from being an iPad-sized hit and the goal of its competitors was far from becoming that.
It seems to me like AyaNeo, GPD and the rest did what they set out to do: boutique handheld PCs that meet the goals of the crowdfunding platforms they use to get them made. And eventually Valve went "hey, there's something to this, maybe we can return to the Steam machines idea in this space" and they did that, too.
@darth@MudMan@jbzfn We're getting off a bit on tangents here. The whole point is just proper crediting. Valve didn't make the APU, just asked AMD to give them something to certain specs they already could meet and AMD made it. They didn't make the formfactor (basically a copy of others that predated it including the Switch) or the control design.
But they did put it all together, make it viable for gaming, priced it well, and made it durable enough.
@darth@MudMan@jbzfn@euronews I get so paranoid about harddrives. Officially even the best ones are only guaranteed for up to 5 years. Most were 3. I had an old Maxtor in my PS2 that was going on 12 years or so and I got paranoid and finally switched it to a SSD via adapter. No more clicking and etc sadly, but then the PS2 wasn't know for HDs anyway, lol.
@nazokiyoubinbou@darth@jbzfn Yeah,. I have much worse luck with hard drives. I'm always impressed when retro collectors keep their old drives working on XTs and ATs and whatnot, but it always seems like a matter of time until it spontaneously combusts.
I'm weirdly less concerned about hybridization there. Maybe because hard drive noises were less consistent, so less iconic. Maybe because the change was more gradual. Maybe because I still use HDDs in modern mass storage.
Add comment