Supercard vs Kryoflux

[dlfrsilver]

Originally Posted by desiv

I've heard this type of thing mentioned before, but I've had really good luck with my floppies.. Much lower than 30% bad...
And I seem to have even better luck with my older 5 1/4" floppies...
(knocking on wood)

That said, I would recommend Kryoflux for me...
Because I bought one (before there was a Supercard).

desiv

indeed, that's more or less what i have seen too, in my own collection.

I have Lorna for PC on 5.1/4, viaje al centro de la tierra too, Game Over II and Narco Police in this format.

I have been able to image most of them without any errors. The only problem is that they are copy-protected with weak-bits/weak Sectors protections.....

[JimDrew]

Originally Posted by Jope

Also, if we consider an ideal imaginary perfect signal path situation and the source signal has a very high signal to noise ratio, increasing the sample resolution a bit like taking a bitmap and scaling it 2x or 4x. You get bigger pixels, but you don't get any additional information.

You have this backwards. We are down-scaling, not up-scaling. Extra information provides substantially more accurate end timing for the decoder, otherwise we could just sample every 1us and it would be fine, which is not the case. The limit seems to be right at 10ns, with any captures faster than that offering no benefit.

As I stated before, the SuperCard Pro software only has analyzing support for Amiga disks, which can be preserved as .adf images. However, there are several other programs that can use the SCP hardware to read/analyze and image or copy disks of various disk formats. There are no other programs that use the Kryoflux hardware to read/analyze/write disks because the Kryoflux information is deliberately withheld from the public, and that is the very reason why I created SuperCard Pro and made the information for the image file format and control of the hardware available to everyone. This is why there are numerous programs and emulations that support the SCP platform.

[Jope]

You have this backwards. We are down-scaling, not up-scaling. Extra information provides substantially more accurate end timing for the decoder, otherwise we could just sample every 1us and it would be fine, which is not the case. The limit seems to be right at 10ns, with any captures faster than that offering no benefit.

No, I meant what I meant. A signal of "A" sampled 10 times in the same amount of time is "AAAAAAAAAA" in an ideal perfect world. Now if your data source is not perfect, perhaps your end result might be "AAaAAbABAA" instead. Will this bring us benefit? I will leave this to the audience to think about, your opinion is visible above. The disks we are imaging were mastered at a finite resolution. It does not provide much more than marketing benefit to increase the sampling frequency far above what was written onto the disk.

Originally Posted by JimDrew

However, there are several other programs that can use the SCP hardware to read/analyze and image or copy disks of various disk formats. There are no other programs that use the Kryoflux hardware to read/analyze/write disks because the Kryoflux information is deliberately withheld from the public

I don't believe kryoflux is any more closed than your product.
https://www.kryoflux.com/download/kr...col_rev1.1.pdf

Anyone is free to implement support for uncooked streams in their product and even create an ipf file out of that data without any checking/massaging beforehand, resulting in something that is just as useful as a scp dump. The ipf format has support in many emulators.

ipf files created by the software preservation society go though a long manual analysis step to ensure that they will actually result in the same kind of disk when mastered back. Homebrew ipfs blindly created using the dtc tool's converter often work, but may not always result in an identical master disk when written back to floppies, if the disk format is tricky enough.

As for accessing the hardware without the dtc tool, I don't know whether that is open or whether it is useful to have. Perhaps it is. This is down to the end user to decide. I am certain, that if someone would run into the limitations of the stream fileformat, they are fully qualified to make the decision between these two products themselves.

The most exciting thing I've seen that was dumped from stream images was a Quantel Paintbox floppy with no sector structure. https://forum.kryoflux.com/viewtopic.php?f=3&t=1310 limits not found yet!

I don't see the need to revise what I've written otherwise. I will leave it down to the readers to read through the postings and make an educated decision about which product is most useful for them and for which reasons.

[tomse]

Well Keir Frasers Disk-Analyze program converts from stream to IPF.
So one can make "unofficial" IPF releases of one's original disks.

while that is good, I had an accident with the Supercard where I had a write protected disk, and accidently copied a d64 file to the disk.. it was a 1541 Demo disk so not a real loss.


@op
You should download each software package and see which one you like the best.

[JimDrew]

Preserving at the highest resolution provides the most accurate preservation, it's as simple as that... otherwise we would all be ripping CD's as 8 bit 22.1KHz samples.

Perhaps you are not familiar with decoding data that has been shifted (magnetically), and how it can be extracted/corrected as the original data. This requires and extremely high resolution sample to get the signal orders.

With the information provided, anyone can control the SuperCard Pro hardware directly to step tracks, turn on/off the drive motor, read/write flux, etc. and a USB port is not required for controlling it. You can not control the Kryoflux hardware directly as that information is not publciy available, and there is no way to access it by any means except USB.

The .scp image file format is also open for everyone to use, and because of this there are many programs that support it. If you feel compelled to use the Kryoflux image format you can use the HxC Floppy Drive Emulator software to convert between .raw and .scp formats. Keir's utils can also control the SuperCard Pro hardware and output in various formats.

[Ratteler]

Has any one tried these with a 2.8MB Drive? I bought one for maximum compatibility with all media.

[demolition]

Originally Posted by Jope

No, I meant what I meant. A signal of "A" sampled 10 times in the same amount of time is "AAAAAAAAAA" in an ideal perfect world. Now if your data source is not perfect, perhaps your end result might be "AAaAAbABAA" instead. Will this bring us benefit? I will leave this to the audience to think about, your opinion is visible above. The disks we are imaging were mastered at a finite resolution. It does not provide much more than marketing benefit to increase the sampling frequency far above what was written onto the disk.

Oversampling the source is not generally a negative thing. At some point you are just increasing the noise and not the signal, but you are also not losing any signal and you can remove the extra noise by properly low pass filtering the signal.

[prowler]

I have nothing useful to add to this thread. The comments made by other respondents show that they are more experienced than I in using these tools.

[mr.vince]

The KryoFlux is perfect for all things floppy and the superior product. It's already oversampling the data many times, and while one can still oversample it even more that's not going to give more information.

Raw stream data format is very well documented and supported by a vast range of third party tools.

Check out who's already using the KryoFlux:
https://kryoflux.com/?page=kf_refcustomers

As for the rest, Jope's explanations are pretty comprehensive.

The KryoFlux comes with a 14 day return option. If you're not satisfied, you give it back for a full refund. For the record: that's EU law.

[McTrinsic]

Hmmm, both developers of competing products have arrived.
I'll go and get some popcorn. This will be fun.

[JimDrew]

SuperCard Pro is being used by dozens of universities, the U.S. library of congress, the Smithonian Institute, and other organizations. The cost of SuperCard Pro is the same to everyone, including professional organizations. Thousands of SuperCard Pro boards have been sold, and I am now producing the 6th production run.

[idrougge]

Originally Posted by desiv

I've heard this type of thing mentioned before, but I've had really good luck with my floppies.. Much lower than 30% bad...

I have a 33 % failure ratio when trying to format floppies nowadays…

[ptyerman]

I own neither at the moment so don't have much say in the matter, but, I would think it depends on what side of the Atlantic you are. Personally, being in the UK it's the KryoFlux I will eventually go for I think, if I was in the Americas it would be the SuperCard.

[tomse]

Originally Posted by McTrinsic

Hmmm, both developers of competing products have arrived.
I'll go and get some popcorn. This will be fun.

get the big screen ready

[sean_sk]

Originally Posted by demolition

you can remove the extra noise by properly low pass filtering the signal.

That's very interesting. How would you go about doing that? What software could you use?

[JimDrew]

You actually use a bandpass filter tailored to specific disk format you are looking at. There are many open-source examples that support Supercard Pro's 25ns capture resolution. Take a look at Keir's disk utilities source code.

[NLS]

I see this thread just "stopped" a while ago.
Yet, I am thinking of getting one of these things (and I am already late)...

So question is: Any new data to enrich the thread?
Actually HAVING new data (for either card) to enrich the thread is already a good indication of what card to choose. It is called "support your product and keep it current" (which ok is a bit weird to say about retro stuff, but I am sure there is always debugging to be done and possibly new ideas to implement).

So... ANY NEWS? When are the latest versions of each of these developed?

[demolition]

I think they both work well for what they were designed for so nothing major has happened with either of them. Why fix something that works?

[tomse]

you can find software updates on the Kryoflux website, and new versions come out every now and then, last release was less than a year ago. For supercard pro the same.. so they both seem active and still supported.

but handling of the stream (kryoflux) files from the softpress team is another matter.. still waiting for answers after 6 years.
Which is of course depending on your question.

[demolition]

There have also been regular improvements to the software that you use to process and convert the raw dumps, for example disk-analyse and HxCFloppyEmulator.