ICD Multi I/O

[danwinslow]

Fully populated, and tested functional.

 

In that case I'll take 2

 

Ok. I'll run them down the street to you, as soon as they're done. hahah.

 

But seriously, I'm estimating another week on the board layout. I'll make an initial run of 10 boards. When I get one asembled & tested, I'll post a new topic here in the forums, and start taking pre-orders. As soon as I get the money for the first 10, I'll order all the components and assemble/test the rest of them and ship them out.

 

Then from that point on, it just depends how many more people want one.

 

Im gonna be dead honest. This is a SH*TLOAD of work. I'm probably going to be making about $40.00 over parts/materials cost per unit, and we are talking about somewhere in the neighborhood of 500 solder joints per unit (a good bit of assembly time). I have devoted every spare moment of free time to this for the past 2 weeks, and Ive got at least a week to go just on the board layout, Im estimating. All of the money I get above the cost of materials for the first run will be spent on parts for the second run. I'm not going to see any compensation for my labor until Im well into the second production run.

 

I do this COMPLELTELY for the benefit of the ATARI community, and in the end, maybe we'll at least have a bit larger user base for the MIO.

 

When I was a kid, I wanted one of these so bad, I'd have given my left nut for one. Now maybe a few more of us can have one, at a reasonable price.

 

Ken, just so you know, I'd be up for buying all of them you have left if you have trouble moving all 10. I don't think you will, but thought I would mention you can count on moving them all. Also, if some additional investment would help you ramp up ( by getting socketed boards or something ) let me know the particulars and we could talk.

[+remowilliams]

Fully populated, and tested functional.

Put me down for one

[MEtalGuy66]

Yeah. EVERYTHING is defintely gonna be socketed. I cant stand companies that built thru-hole boards without sockets. I mean we are talking like a dollar extra to put sockets. I dont have a problem delsoldering thru-hole chips to do repairs, but WHY???

 

Maybe its the ex-mechanic in me, but I feel that any design thats worth a crap should be servicable, and servicablility is a major factor in how good any design is.

 

That's very kind of you to offer to help finance it, Dan. As long as I can get pre-orders, I think everything should work out just fine.

[Almost Rice]

Yeah. EVERYTHING is defintely gonna be socketed. I cant stand companies that built thru-hole boards without sockets. I mean we are talking like a dollar extra to put sockets. I dont have a problem delsoldering thru-hole chips to do repairs, but WHY???

 

Maybe its the ex-mechanic in me, but I feel that any design thats worth a crap should be servicable, and servicablility is a major factor in how good any design is.

 

 

You already have the sale. Thats my position electronics too. I have an aversion to Tramel era 8bits because of the lack of sockets.

Edited April 9, 2006 by Almost Rice

[MEtalGuy66]

Ok. I am just about finished with the initial design work on the board layout for the new MIO. What I am doing now is meticulously checking every single trace and component placement to make sure everything is 100% right. This is a slow process and Im still looking at many hours of work left, but this will hopefully eliminate any errors or the need for any "post production fixes"..

 

 

DOES ANYONE HAVE A 1 MEG MIO AT THEIR DISPOSAL?

 

If so, I need to get in touch with you, and hopefully get you to check a few things on the front and back side of the board. This is because the only actual MIO board I have to work with is a 256k model, and in designing the 72pin SIMM ram interface on the new MIO, there's a few things I need to know for sure, in order to guarantee 100% compatability of the new design with the original 1 meg MIO, where the RAMdisk/Print-spooler cofiguration is concerned.

 

If someone (who owns a 1meg MIO) could help me in this regard, I'd be VERY grateful.

 

You can email me at kjones66@earthlink.net

 

or just PM me on this forum.

[MEtalGuy66]

If someone (who owns a 1meg MIO) could help me in this regard, I'd be VERY grateful.

 

You can email me at kjones66@earthlink.net

 

or just PM me on this forum.

 

Ok. Dan has graciously offered to disassemble his 1 meg MIO unit and send me high res pics of both sides of the board, so that should be taken care of.

 

Thanks, Dan.

[carmel_andrews]

Was ICD's MIO ever released in EU/UK... I never saw one and also I don't recall Frontier software (ICD's main EU/UK distributor) stocking or promoting it, if that isn't bad enuff, i don't recall Sillicia shop (EU/UK's Largest Atari related dealer/distributor) even bothering to advertise or stock it

 

I guess that there wasn't a market for it in europe or the UK...shame

[Mathy]

Howdy Carmel, folks

 

Was ICD's MIO ever released in EU/UK...

I saw one at an Atari Fair in the Netherlands years ago. Some Dutch company was selling it. At the moment, I didn't have the money for it. Later the BlackBox appeared and I choose to buy that one. Only thing it doesn't have is huge amounts of RAM like the MIO, unfortunately.

 

Just out of curiousity, would it be possible to expand the memory of a 1MB MIO? (just curiousity, I'm not interested in buying a MIO)

 

greetings

 

Mathy

[+warerat]

Howdy Carmel, folks

 

Was ICD's MIO ever released in EU/UK...

I saw one at an Atari Fair in the Netherlands years ago. Some Dutch company was selling it. At the moment, I didn't have the money for it. Later the BlackBox appeared and I choose to buy that one. Only thing it doesn't have is huge amounts of RAM like the MIO, unfortunately.

 

Just out of curiousity, would it be possible to expand the memory of a 1MB MIO? (just curiousity, I'm not interested in buying a MIO)

 

greetings

 

Mathy

 

Don't see why not. The RAM is mapped at $D600-$D6FF so you have to bank stuff in 256-bytes at a time, addressed by writing to two latches. You'd have to map another latch somewhere at $D1E4-$D1FE.

 

I'm already thinking about using the unused 8K in the MIO ROM. They were supposed to hold the 80-column code. I'd like to have IDE routines in there instead, and map a 16-bit interface in at $D180 or so. It's a really neat interface.

 

I don't have any schematics so I don't know if the logic is there to use the remaning four 2K banks or not, if not, it'd be easy to add to the existing latch.

[MEtalGuy66]

ICD may have had the design for the 80 column board complete, but at the time of actual production of the MIO, they already knew they werent going to follow through and produce it. I say this because on the MIO boards, they never actually populated the holes for the header pins for the RGB connector, or the second 50 pin row of header pins that was in paralell with the expansion bus connector. These are undoubtedly the places that the board would have pluged in. They chose instead to fill these holes with solder. This would have required all MIOs to be shipped back to ICD for installation of the 80 column board (they would have had to desolder all those holes, insert the headers, and resolder them).

 

The boards Im making will have the ALL headers (including the above mentioned ones, and the auxilary power header near the rectifier/regulator circuit) populated with pins. This way, if someone manages to unearth the design for the 80 column board at some future time, we can build it, and just plug it in. In the mean time, Dan Winslow and I will be working on an ethernet/firmware TCP module that plugs in there. He has already done quite a bit of work and already has the TCP stack running on the module's imbedded processor, and passing packets to/from the atari via SIO communications. We will be adding thenecessary interface logic, if possible, to make this an internal expansion for the MIO via the full 8-bit data bus, which will hopefully produce very useable throughput rates.

 

WereRat, I have the MIO schematic if you need it.

Edited April 12, 2006 by MEtalGuy66

[mos6507]

There are two problems with the MIO vs. the Black Box.

 

1) it only allows drives with 256 byte sectors. BB allows 512 byte sectors.

2) the serial port doesn't support hardware flow-control.

 

For a modern recreation of the MIO, you'd really want to resolve both of these limitations.

[carmel_andrews]

Which was better... Black Box (which sounds kinda super duper) or the MIO (which looks a bit basic)

 

Never owned either, so have no exp. of using these items

[MEtalGuy66]

There are two problems with the MIO vs. the Black Box.

 

1) it only allows drives with 256 byte sectors. BB allows 512 byte sectors.

2) the serial port doesn't support hardware flow-control.

 

For a modern recreation of the MIO, you'd really want to resolve both of these limitations.

 

Nope. I DON'T wanna "resolve" either of those issues.. 256 byte sectors are fine for an 8-it computer, and Ive got a freakin STACK of old SCSI hardisks that will work perfectly with it. Also, as far as SPartados compatability is concerned (which is the only dos I really care to use on an ATARI), there is no better supported interface than the existing MIO design. WHat I want is 100% compatability with the original MIO design and all fo the software that has been written to utilize it. I really dont care about sector-size, or wasting space on partitions as a result. The atari can only deal with 16 meg partitions as a maximum anywayze, and every piece of decent software ever written for the atari will fit in a couple hundred megs. If you want "modern" hardisk interface, then shell out the cash for one of the more recently developed designs such as the black box.

 

As far as the serial ports are concerned, I dont see why it doesnt support hardware flow control. Its driven by a 6551 uart chip, which includes RTS, CTS lines. This is much better than the 850 interface/ P:R: connection, which are not UART driven. IF in fact it doesnt support hardware flow control, then I'd suspect that to be a software issue. Im not oposed to an udated ROM image for the MIO, but Im not changing the actual design spec, because It is the closest thing to a STANDARD that has ever existed for ATARI hardisks, especially where Spartados is concerned, and I want the option of retaining absolute 100% compatability. If compatability isnt important to you, then why not go with MyIDE? Its alot cheaper.

[+warerat]

There are two problems with the MIO vs. the Black Box.

 

1) it only allows drives with 256 byte sectors. BB allows 512 byte sectors.

2) the serial port doesn't support hardware flow-control.

 

For a modern recreation of the MIO, you'd really want to resolve both of these limitations.

 

Nope. I DON'T wanna "resolve" either of those issues.. 256 byte sectors are fine for an 8-it computer, and Ive got a freakin STACK of old SCSI hardisks that will work perfectly with it. Also, as far as SPartados compatability is concerned (which is the only dos I really care to use on an ATARI), there is no better supported interface than the existing MIO design. WHat I want is 100% compatability with the original MIO design and all fo the software that has been written to utilize it. I really dont care about sector-size, or wasting space on partitions as a result. The atari can only deal with 16 meg partitions as a maximum anywayze, and every piece of decent software ever written for the atari will fit in a couple hundred megs. If you want "modern" hardisk interface, then shell out the cash for one of the more recently developed designs such as the black box.

 

As far as the serial ports are concerned, I dont see why it doesnt support hardware flow control. Its driven by a 6551 uart chip, which includes RTS, CTS lines. This is much better than the 850 interface/ P:R: connection, which are not UART driven. IF in fact it doesnt support hardware flow control, then I'd suspect that to be a software issue. Im not oposed to an udated ROM image for the MIO, but Im not changing the actual design spec, because It is the closest thing to a STANDARD that has ever existed for ATARI hardisks, especially where Spartados is concerned, and I want the option of retaining absolute 100% compatability. If compatability isnt important to you, then why not go with MyIDE? Its alot cheaper.

 

On your schematic, are CTS, DSR, and DCD tied to ground?

 

The reason I'm buying one from you is because I'd like to enhance it while retaining 100% compatibility. Thanks for leaving the extra 50-pin header in place, I'm going to need it.

[Mathy]

Howdy Warerat

 

I'm already thinking about using the unused 8K in the MIO ROM. They were supposed to hold the 80-column code. I'd like to have IDE routines in there instead, and map a 16-bit interface in at $D180 or so. It's a really neat interface.

 

Maybe you can put an ASPI driver inthere. See my ASPI page.

 

Tools written with ASPI support work on all interfaces that support ASPI. No matter if that's a MIO, BlackBox, SupraInterface, etc. Unfortunately, the only ASPI driver that exists at this moment is the ASPI driver for the BlackBox. But you can see on my site what can be done with it.

 

greetings

 

Mathy (who has controlled/played audio CD's and looked at data CD's and DVD's using his Blackbox)

[Mathy]

Howdy folks

 

As far as I know, the MIO will do hardware flowcontrol. But IIRC you need a piece of software to enable it. Sorry, can't remember the name.

 

Here's my list of pro's and con's for both. I personally own a BlackBox plus FloppyBoard and have never owned or used (but have drooled over many articles and adds mentioning the MIO), so I might be biased.

 

MIO:

- has a very nice case

- switch in the menu and the code that was in RAM at that location is gone.

- has lot's of memory

- 80 columns were planned but never released.

 

Blackbox:

- A very ugly case is/was an option

- No memory. But a printer spooler (64kB) can be bought or build with parts available at the local electronics store.

- The menu can be switched in safely. Over 95% of all software will run as usual after leaving the menu.

- FloppyBoard (an option) allows you to connect 4 PC floppy drives.

- 512 byte sectors can be used. They are "cut into pieces" so DOS only sees 256 bytes at a time. No bytes are lost.

- 96 partitions can be loaded into a partitionlist. But we've (read: Matthias Reichl and maybe me) come up with a trick, so you can have as many partitionlists on your HDD as your HDD (or ZIPdrive, or CD or DVD) can handle. If somebody wants to right a tool for switching partitionlists, please go ahead. Your work will be appreciated.

- 80 columns could be added in a simular way as the FloppyBoard is added, but somebody has to develope this upgrade first

- A PC Tool to create boot CD's has been developed. The idea is mine, the software is Matthias Reichl's. Believe me, not all software will fit on one CD. I have a CD that is half full and only contains ABBUC disks (mags, PD) and AMG (mags, games).

- The configuration data (as saved to sector zero of your HDD) has be decoded (by me) and can be found on my BlackBox page (it's at the bottom of the page).

 

If somebody decodes the stuff other HDD interfaces write to HDD, the PC Tool (for creating boot CD's) can be made to support different HDD interfaces too.

 

Greetings

 

Mathy

[+warerat]

WereRat, I have the MIO schematic if you need it.

 

I'd like to take a look at them...

[ddez]

ICD may have had the design for the 80 column board complete, but at the time of actual production of the MIO, they already knew they werent going to follow through and produce it. I say this because on the MIO boards, they never actually populated the holes for the header pins for the RGB connector, or the second 50 pin row of header pins that was in paralell with the expansion bus connector.

 

Are you sure about that? Here's a pic of my 256K MIO, it has both headers installed.

 

ddez

 

[mos6507]

As far as the serial ports are concerned, I dont see why it doesnt support hardware flow control.

 

You should try one and see how well it runs with a 56K modem attached. The Black Box will throttle down the effective throughput to the maximum 19.2K the port allows. I don't think the MIO will handle it without manually limiting the modem to something like 14.4 with data compression disabled.

[ddez]

IF in fact it doesnt support hardware flow control, then I'd suspect that to be a software issue.

It's a software issue, here's a replacement R: driver that supports hardware flow control:

 

http://members.aol.com/lenspencer/hyperspd.htm

 

ddez

[MEtalGuy66]

Heh. Well first of all. if you lok at the specs of the 6551 UART, it's max baud rate is 19.2k, so I wouldnt expect it to do 56k at full speed. The 6551 was designed originally by MOS technologies, specifically for use in 6502 based systems, and the advantage of the UART is that it takes the load off of the CPU, where serial communications are concerned. You have to remember though that the 6502 was designed in the 70s, and 56k, at that time, was unheard of, (hell, a 300bps modedm would run you hundreds of dollars.) I could change the design to use a later Intel 16550 UART, I think they are capable of speeds over 150kbps.. But in the interest of compatability, I am producing this board as a 100% functionality clone of the original MIO. Everything on it will be socketed, so theres all kinds of possibilites of making upgrades via "daughter boards" that plug into the chip sockets. I just want first and foremost, 100% compatability with the existing ICD/FTE software and OS to begin with... From there, we can start designing upgrade mods.

 

 

Thanks for the pic of the board with the populated pins. If you notice, that serial number is in the 200s. Both of the boards I have seen (that didnt have the pins) were numbered in the 700s.. So I guess that proves that initially, they did plan on at least the possibility of releasing an internal expansin board (80 column board or otherwise). But at some point, they decided to start building the MIO without the expansion headers installed.

[MEtalGuy66]

The serial and paralell ports on the boards I am making will have the pins/gender rearranged so that they are 100% signal compatable to standard PC serial and Paralell ports. The serial port will be a male DB9 (instead of female db9 on the original MIO) and the paralell port will be a female db25. standard PC serial/modem, and printer cables will work with it.

 

progress report on MIO clone board design:

 

precision mesurement and dimensinal positioning of components (to +/- .010") is 100% complete.

 

silkscreen layer is 100% comlete.

 

Major "beef-up" of power-supply stage design is 100% complete.

 

Redesign of RAM network circuit (to accomadate 72pin SIMM socket, instead of multiple SIPP components) is 100% complete.

 

bottom copper layer logic traces is 100% complete.

 

bottom copper layer "double checking" (one trace at a time) is 100% complete.

 

top copper layer logic traces is 50% complete.

 

Still to go:

 

finish top copper layer logic traces.

 

"double check" top copper layer, trace by trace.

 

make necessary "rerouting" to convert serial/paralell ports to PC standard.

 

"double check" RAM circuit design (for compatability) with info recieved from Dan's inspection of his 1meg MIO.

 

"Triple Check" EVERYTHING!

 

With any luck, I should be ready to have the boards made by monday or tuesday.

[+Larry]

> ...and Ive got a freakin STACK of old SCSI hardisks that will work perfectly with it. <

 

Interesting. What SCSI hard drives do you have that will accept 256-byte sectors? Or have you "reprogrammed" the drives internal code to allow 256-byte sectors with something like "Bart's SCSI Tool?" I ask because I've only found one SCSI drive that allowed formatting with "small" sectors a Seagate 225N. MFM with a bridge-board (the latter now really hard to find), no problem.

[+Larry]

> The MyIDE (original external cart version with 3.1 drivers) is a dog by comparison:

DOS writing: 2806.680942 B/sek.

DOS reading: 2849.391304 B/sek.

DOS average: 2828.036123 B/sek. <

 

Here are some recent figures for the MyIDE flashcart V3.5f used with a compactflash card:

Dos Writing 11988

Dos Reading 9781

Dos Average 10885

 

Quite a bit improved from the earlier figures posted here. I suspect that these figures can be influenced quite a bit by how full/fragmented the HD is. For the best measurement of the interface's potential, you'd want to use an essentially blank HD. Incidently, in the tests that I've run with the KMK/JZ interface, there is very little difference between compactflash and HD figures -- at least on "empty" drives/cards.

[drac030]

It is expected that you run the test on a blank partition to reduce the influence of the DOS overhead searching for free sectors to be allocated for the file.