KryoFlux Public Beta 1.0 - AVAILABLE NOW!

[mr.vince]

Originally Posted by Supamax

I see that the board supports two floppy drives...
how does it make use of 2 drives? Are they controlled separately (for example a 3.5" and a 5.25" for different source floppies), or can they be used at the same time, for example for direct floppy copying purposes (original ---> empty floppy copying without disk swapping) ?

In other words, how will exactly Kryoflux make use of the two drives?

KF can see two drives on the bus, like an ordinary FDC in a PC can. So yes, KF can switch between these. How they are used depends on the software. Our idea was to enable users to permanently have two drives attached for dumping of 3.5" and 5.25" disks.

It would be possible to write software that does 1:1 copying, but this will be the same like on any other bus. 1. read. 2. write. You can't access both drives at the same time due to the fact that this is a bus. So yes, two 3.5" drives can be hooked up as well. This depends on the cable of course.

In the current scenario, DTC will dump raw data or a sector dump to disk. A second piece of software would then have to pick up this data, transform it to something else (like the Analyser does for IPFs) or write it back to disk.

But again, this depends on the software being used.

Cheers! Sommer's outside. Yay!

[gizmomelb]

reading and archiving old floppy discs is a very fine thing, however writing back to floppies isn't quite as useful any more (or for the future).

http://mdn.mainichi.jp/mdnnews/busin...na008000c.html

a solid state floppy emulator with write support is going to be what is needed for retro fans to keep their vintage computer equipment operational.

[mr.vince]

I did post that news yesterday on the Softpres Facebook page. DD disks, as needed for the Amiga, went out of production 10 years ago. But to be honest... how long has it been that 5.25" disks were officially abandoned? 15 years? 20?

http://www.athana.com/html/diskette.html

I think they are still in business.

If you're really scared because disk production was halted, I'd recommend getting rid of your Amiga as well. Production was halted over a decade ago, so what you really need to keep the spirit alive is an emulator.

BTW: ever heard of HxC?

[fiath]

Apologies for being tardy, there have been three pieces of news I didn't post here yet:

http://softpres.org/news:2010-04-20
http://softpres.org/news:2010-04-27
http://softpres.org/news:2010-04-28

[gizmomelb]

Originally Posted by mr.vince

If you're really scared because disk production was halted, I'd recommend getting rid of your Amiga as well. Production was halted over a decade ago, so what you really need to keep the spirit alive is an emulator.

that's pretty much the road I'm going now. the hardware is getting too dodgy and will die, the disks are unreadable more often that not and the emulator is getting more and more accurate and is much faster and easier to use with virtual floppies (which won't die due to bit rot or mechanical failure).

Originally Posted by mr.vince

BTW: ever heard of HxC?

Strangely enough, I have. I posted first here about it some two or more years ago. Like Kyroflux it is an excellent hardware project. However also like Kyroflux it doesn't support writing to Amiga floppy images.

[mr.vince]

Originally Posted by gizmomelb

that's pretty much the road I'm going now. the hardware is getting too dodgy and will die, the disks are unreadable more often that not and the emulator is getting more and more accurate and is much faster and easier to use with virtual floppies (which won't die due to bit rot or mechanical failure).

So you understand there needs to be a device that can at least READ data off floppies properly. Writing is only a bonus.

Originally Posted by gizmomelb

Strangely enough, I have. I posted first here about it some two or more years ago. Like Kyroflux it is an excellent hardware project. However also like Kyroflux it doesn't support writing to Amiga floppy images.

Don't worry. Things will evolve. Besides, I bet Jeff is happy for someone to come along and offer a helping hand!

[Loedown]

Originally Posted by mr.vince

So you understand there needs to be a device that can at least READ data off floppies properly. Writing is only a bonus.

No, writing is also a neccesity, not everything can be done in the virtual realm, I don't feel like spending bikkies on Kryoflux and HxC too. Just the routines you guys said, the coding.

[mr.vince]

You did not understand what I wrote. To preserve data, you must be able to 1. read. That's what we're after in the first place. When you can't ingest, you can't process or write back anything.

So writing, from our point of view, is a bonus. It does not help us do our work from a preservational point of view. Nevertheless, we've planned write support for the future.

[Loedown]

Originally Posted by mr.vince

You did not understand what I wrote. To preserve data, you must be able to 1. read. That's what we're after in the first place. When you can't ingest, you can't process or write back anything.

So writing, from our point of view, is a bonus. It does not help us do our work from a preservational point of view. Nevertheless, we've planned write support for the future.

English is a great language and very ambiguous, whilst I understand it's vital to be able to read the data correctly in the first place, writing is also important for the reasons I already specified. Writing for the future, one perhaps should not make it too distant.

[Pheonix]

Preservation is only usefull as long as the material can be used. I would say that writing back would be necessary, only if there is no other way of using it that is 100% compatable.

I don't know much about the HxC project, I stumbled across it once, it sounded interesting, but after 2 hours of searching, I gave up being able to buy one. Because I cannot seem to get it, preservation that depends on it in order to use the material is absolutely useless to me. Also, I don't know if it is completely compatable. Are there some programs that won't run through it?

The next option (without write back capability,) is to only use it in emulators. Well, I've been very impressed with the emulators that are out now, but are they completely 100% and absolutely compatable? I, personally, don't think they can ever be, without building hardware with original chips & such for the emulator to fall back on. I'm willing to admit that I'm wrong on that last point, but remember, I'm talking about absolute 100% compatability and functionality (not 99.9% or almost complete, but 100%.)

Finally, there are nostalgia purists out there, who will miss the buzz-click of the drive working (yay to UAE - or maybe just WinUAE - for adding in the sound effects.)

The only way to have the complete and 100% compatable experience would be with original hardware, including the floppy drive, with physical magnetic media floppy disks, and not a RAM card you stick in a memory slot.

So, from my point of view, it seems that for preservation to be usefull, there must be write back capability. I have a game, I want to preserve it for when the floppy finally fails (which it will some day.) I make an image of the disk(s). Now, it finally fails, after 2 weeks, 2 years, whatever. However, I have this nice image I preserved for this day. Only, if "I" cannot do anything with the image except look at it sitting on my PC, what good is it? Well, I can take it to a friends house and play it on his PC, if I can talk him into loading an emulator (they won't run on my 486 sx33 - Heck, they won't run on my P2-233 system either).... Ok, that's not an options. Next, I see this nice fake floppy that will run the image off a SDRAM card (I have some of those for my camera.) Only, I cannot get one as nobody sells them any more. So, there goes that option. So I'm left with this nice image that sits on my PC, not even looking pretty.

I don't mean to be negative, but I personally feel that writing should be a major project point. Not something you plan to maybe do, but something you are deffinately going to do. It should be a deffinate step 2. Step 1 - Be able to read reliably & accurately, the raw data. Step 2 - Be able to reliably & accurately write that raw data back to a disk and the disk work. There, now you have a solid preservation method that fulfills the function of preservation - that is to save something perishable in a usable manner. Step 3 - (label this as a bonus if you want,) be able to interpret the raw data into a form usable by an emulator be it hardware floppy emulation, or software system emulation. Over simplified, I know, but this is just the basic project points, not actual implimentation here.

I don't personally think #3 should be a "bonus" either, but for the sake of preservation it is not absolutely necessary. I also don't see any problem with working on #2 & #3 simultaneously as they are different enough that they don't have to be done by the same team. Ok, don't know much about the SPS, are there enough people to have broken things down into teams? Push comes to shove, I personally don't see a problem with doing #3 before #2 either, as long as #2 comes along eventually (preferably very quickly.)

[IFW]

Due to the way magnetic recording works, you absolutely and positively have to understand what you read in order to be able to create data that can be written and later read to the exact same effect without degradation in quality.

It's a very long topic, but to make a long story short: writing is the bonus of a very complex analysation process and reproducing mastering data. Anything less is just hacking and depends on luck.

Preservation is not about copying regarding digital artefacts.
It is reading data, understanding data, and being able to create the exact same data. You'd also have to care about data integrity and data authenticity.

This is much more complex than most people would think, especially for floppy disks, where modifying the original data by consumer grade equipment is a possibility.

So yes, writing is planned, but that is a byproduct of all preservation efforts. Writing without the preservation is easily achievable - but that's not very useful.

[IFW]

http://softpres.org/knowledge_base

These articles are good introductions to the topics referred:
http://softpres.org/article:importan...data_integrity
http://softpres.org/article:importan...a_authenticity

[gizmomelb]

Originally Posted by IFW

So yes, writing is planned, but that is a byproduct of all preservation efforts. Writing without the preservation is easily achievable - but that's not very useful.

a writeable floppy emulator (for saving high scores) is what would interest quite possibly 99% of retro computing fans and would easily fund the development for the perfect 'bit archiver' which would then satisfy the other 1% of the retro computing market.

it will happen, whether it's HxC or Kyroflux or someone else.. and that developer stands to make a nice amount of money in a niche market.

Every time there is a production run of the HxC (USB or SD) or any of the other similar floppy emulator devices (usually priced around 60-100 Euros), they sell out very quickly (usually batches of 50 or so) - usually with the only 'advertising' being mentioned in the developer's forum or a few times now being listed on ebay.

[Supamax]

Originally Posted by gizmomelb

it will happen, whether it's HxC or Kyroflux or someone else.. and that developer stands to make a nice amount of money in a niche market.

You are right, but this amount of money is usually enough to cover production costs and not (much) more.
They are made with passion and with preservation in mind, not in order to make money.
So I'm absolutely sure that the "economic" point of view is not related at all to projects like this .

[mr.vince]

Originally Posted by Supamax

So I'm absolutely sure that the "economic" point of view is not related at all to projects like this .

Softpres was not founded because of money. However, whole project has grown beyond a pure hobby, with expenses here and work there. A lot of things is happening behind the scenes, like talks with other libraries or museums or evaluation of our processes and technology for EU projects.

This all amounts in time (and money) invested. And we're not talking a free hour here and there, but thousands of hours in manpower over the last couple of years. It's easy to say you're doing this because you like to do it. But when you're starting to invest days of vacation, or even leave for a day - unpaid - things look different.

I am pretty sure that when we'd have to make a quote for the KryoFlux project and we would calculate like a company, our work invested (not counting the proof of concept by Rich Aplin, but also counting in the time spent on the hardware design & layout, drives and additional hardware for testing, the logo creation, making photos, writing of the manual, packaging, logicstics) would easily lead to a figure with five zeroes. Do we expect being compensated for this? No.

So why are we doing it? Well, we still think it's important and it needs to be done. And since there's no one else (and it's not meant to sound arrogant) who can do this at the same level of quality... there's no way to stop. Otherwise the last nine years of preservation are wasted.

So to come back to where this started... yes, writing will come. But it will happen as a bonus, the logical consequence. But it does not have priority right now. First, ingestion must work (which, looking at the results of the beta program, is doing pretty well), then we need to transform the RAW data to something (means adding many more formats to the Analyser that does IPFs, like generic MFM, FM, GCR - effectively everything non-Amiga related) and after that we can focus on writing.

We understand writing is important to many users. But we don't have a truckload full of programmers, standing by to do whatever task we give them. Preservation isn't sexy, it's not hot, girls won't come after you and scream "Oh my god, it's him. Remember, he preserved Turrican for the C64 last week". It's ok for us (some of us are even married ), but since it's the way it is, it takes time. More than it should but... it's still just a hobby. More or less... huge.

[Interceptor]

Originally Posted by mr.vince

girls won't come after you and scream "Oh my god, it's him. Remember, he preserved Turrican for the C64 last week"

They don't do that to you?

[mr.vince]

Damn. Must be country specific. I am missing the fun, huh?

[Supamax]

Thanks mr.vince ,
your long and nice post says it all about passion and dedication.
No other words are needed .

[desiv]

An example of preservation that shows that recording is more important that writing, at least at first:
http://articles.sfgate.com/2008-03-2...nal-laboratory

:-)

desiv

[mr.vince]

Your're not expecting them to write back the audio to lampblacked paper, are you?