[Daedalus]

Originally Posted by stefcep2

Nope. That's for the business owner to answer.

I think you missed the bit about the Warp being a hobby project. There's no business currently selling these boards.

Anyway this is a business transaction. He's not emailing them to make friends. Its a simply business query that wold happen countless times everyday and everywhere in the world: "How much and when?"

See above. You're asking people spending some very precious spare time on their hobby to instead spend it answering questions that have already been answered or aren't answerable. If they wish to spend that time, that's great, but it cannot be expected of them.

Bit rude there

I apologise if I've hurt your feelings, but repeating myself for the benefit of this guy when he doesn't understand posts is precisely the sort of thing that might result in curt responses.

Irrelevant.

How?

Who gave him the reasons? Forum posters. Not Good Enough. People selling should give him the reason. In a two sentence email, automated even would be adequate.

See above. Also, they gave responses in various interviews they've given in the community, so it's not coming from forum posters, who are simply relaying the information. I'm also not sure how automating responses would solve the issue.

I think yours a too low tbh. Its not too much to expect a reply email to a sales enquiry from the people selling the damned thing.

Are they selling it? Cool, where can I buy one?

That's the Linux excuse:" Hey this broken or is shit". "Ahh well its not for you them." At least Linux is free

Comparison fail. He's already trying to say why he doesn't want it, all I'm saying is that he doesn't want it.

AN '060 is nice as is RTG. I have them both, But given the price/performance its really an emotional rather than rational decision to buy. And he was emotionally invested, but he's not anymore and that's because his experience with the lack of communication.

Absolutely, that's all pretty obvious stuff.

If its starts with shit service and communication its probably going to end that way.

Indeed, but it's a very different situation when you've bought something (which sounds like it was cobbled together badly from a diagram copied from somewhere) than when you haven't bought something and aren't owed anything.

There's this idea that because we're a niche market, we almost have to BEG to give some sellers our money. Its just not good enough.

This is bizarre - what difference will begging make when there isn't a product to sell? And no due date for the product to be on sale? Begging just makes you look a bit nuts.

Originally Posted by grelbfarlk

I was all set to pay the guy who taped up boxes for Phase 5 for a GREX-4000 when he was all ready to remanufacture them and all I got was a bunch of hentai and some dick pics in response. This was unacceptable service.

Haha, you mean phake 5? That was never a service, and he was never ready to manufacture any products; that was a blatant scam.

Also, I'm detecting a significant amount of ribbon cablephobia from Daedalus and Chucky which is pretty shameful in TYOOLS 2021.

I'm not sure what difference they year makes - do cables under the same conditions radiate less these days? There might be a valuable thesis in that for someone... But I don't have a phobia - I've successfully relocated accelerators on ribbon cables before. I just understand why extending a bus like that is asking for trouble.

Originally Posted by YouKnowWho

I'll just answer you last question here because it is obvious. The reason why drives didn't go parallel is bandwidth.

That's correct, but a very simplistic view of it. Ask yourself *why* parallel bandwidth was limited? Crosstalk, noise and synchronicity are key points, all of which are solved by using a serial link, as PCIe does. The only way they could even get parallel ATA devices to work as quickly as they could was to use specially designed drivers and carefully balanced and buffered buses.

But it didn't stop them from trying, there were plenty of very slow parallel port storage devices done.

Parallel ports are plenty slow enough to not worry about the issues of parallel buses at speed, and the limiting factor was never the parallel transfers but the fact the parallel port was only designed to run at a certain speed. SCSI is probably a better example you should use, though even that went serial.

Actually, one more thing...Blizzard RTG solution is PPC. I'm not interested in the expense of it for one, second not interested in PPC at all, and thus getting it as an RTG solution is significant overkill cost wise.

Indeed, it's a PPC accelerator, but the bus isn't PPC, and neither is the RTG - it's essentially a modified PCI bus. And I'm already well aware you're not interested in PPC, but it's the only example of what we were talking about that was actually released to the market in relatively significant numbers.

Originally Posted by YouKnowWho

NAILED IT! Plenty of this out there.

Where? Why haven't you bought one if there are plenty out there?

What's important to keep in mind also is that there is no high voltage inside the A1200, nothing that these cards normally don't see,

What sort of spike voltages do you get over such a long bus? In terms of noise, reflections and so on, 5V is quite high in itself. And the spikes above 5V might not be the only issue to contend with - there's also undershooting, where the lines actually dip below the ground potential and can cause serious issues, and there's crosstalk, where signals in one wire induce a small voltage in an adjacent wire. The longer the wires are parallel, the more pronounced the effect, and it can creep up enough to push a logical 0 above the lower logic threshold. And there be dragons (and crashes).

and the ribbon wouldn't be travelling over long distance. Nothing more than 8-12 inches at most is needed for flexibility of placement internally.

8-12 inches is very long for extending a bus like that. Consider the clockport, for example. That's a direct attachment to the CPU bus, and you'll struggle to get reliable transfers over an 8 inch cable, even at the slow speeds of the clockport transfers. It could be done of course with very careful design, lots of buffering and termination etc., but it comes back to whether anyone would see it as worthwhile enough to spend the time on it, and a far more sensible and reliable option would be a direct connection to the accelerator as mentioned before.

That Picasso II++ could also be redesigned for purpose to be half height for easy fit in the case with the floppy drive retained. Would be lovely to stack flicker fixer capability on that solution for completeness.

Indeed, along with designing a new accelerator that allows attachment of such a bus. Plenty of stuff to keep you busy on these darkening evenings But, while you're at it, why not do as I suggested and simply add a PCI bus onto your accelerator, like the BlizzardPPC? That way you won't be hobbled by the limitations of Zorro-II, and could use a dramatically more powerful graphics card than the Picasso II++.

The solution would find plenty of users, without doubt.

And how many of those solutions were ever made in large quantities? Were they reliable? Did they work consistently with heavy transfers, like graphics cards? Were they sensitive to interference from other cables nearby? Why aren't the second-hand markets showing plenty of them? Turns out there mightn't be as much demand as you might think. But, as I've said before, stump up the cash and make it back by selling your product to the masses. I'd be happy to be proven wrong here.