The BAD DMA Chip myth

Troubles with your machine? Just want to speak about the latest improvements? This is the place!

Moderators: Mug UK, Zorro 2, Greenious, spiny, Moderator Team

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

The BAD DMA Chip myth

Postby oehansen » Tue Apr 17, 2012 12:33 pm

For the past 5 years, the Atari Community has apparently almost dissappeared, as far as I can see. The Atari having been alive for years, suddenly had a big fallout ... as it was said that almost all ataris were suffering from bad hardware. The DMA chip in the Ataris, all except one ... were labelled "BAD DMA".

It's time to shake this myth ... it's time for some "myth busting".

I made my own device, to read CF cards, and USB "thumb drives". I use a cheap PIC32MX795 based solution, called Max32. I've played around with, and have weeding my bugs out, and playing with this beast for a couple of months. All on my "BAD DMA" Atari. I use HDDRIVER, on the Atari. And have tried MINT, as well as normal atari. I have HDDRIVER boot off the C partition on the drive, and HDDRIVER now recognized the "Thumb drive" as more than 1GB, using group 1 SCSI commands. I even got myself an Atari with a good dma chip, for the sake of safety.

What I did discover, is that the DMA chip is enormously sensitive to timing. There are two different timings on the chip, one for COMMANDS. And one for DATA. You CANNOT use the same timing, for DRQ and IRQ signals. I did the timing in software, rather than with Hardware. I tuned in the signal return, until it worked ... on both the BAD and the GOOD chip Ataris.

So, for all you people out there, that have an Atari with a bad DMA chip. You don't have to throw them away ...

User avatar
spiny
Disk Imager Supreme
Disk Imager Supreme
Posts: 2571
Joined: Mon Aug 11, 2003 11:53 pm
Location: just outside bristol
Contact:

Re: The BAD DMA Chip myth

Postby spiny » Tue Apr 17, 2012 12:41 pm

did you do your tests on an ST or STe ?
The STe is the only Atari with the 'faulty' DMA chip, and only early versions of it. All ST F/FM etc are fine with hard drives.

User avatar
CiH
Atari God
Atari God
Posts: 1134
Joined: Wed Feb 11, 2004 4:34 pm
Location: Middle Earth (Npton) UK
Contact:

Re: The BAD DMA Chip myth

Postby CiH » Tue Apr 17, 2012 12:51 pm

For the past 5 years, the Atari Community has apparently almost dissappeared, as far as I can see. The Atari having been alive for years, suddenly had a big fallout ... as it was said that almost all ataris were suffering from bad hardware. The DMA chip in the Ataris, all except one ... were labelled "BAD DMA".


"SHOCK HORROR!" headline styling aside, are you referring to the early issue STE's perchance?

I had one such machine (still have it), the 'bad DMA' was described by Atari back in the day as only affecting third party hard drives and not 'official' Atari equipment. I found this advice was not wholly correct when I got corrupted directories on an SH204 in a very short time of operation :mrgreen:

The cure was a replacement for the DMA chip, and my old STE is currently running nicely with an UltraSatan, and has behaved well with its predecessor SatanDisk, SH204 and Megafile devices.

But there is a hearty congratulations due if you've found a workaround without needing any hardware replacement. :cheers:
"Where teh feck is teh Hash key on this Mac?!"

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Tue Apr 17, 2012 1:05 pm

All my tests are on a 520STE, with 4 MB installed.

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Tue Apr 17, 2012 1:11 pm

CiH wrote:
For the past 5 years, the Atari Community has apparently almost dissappeared, as far as I can see. The Atari having been alive for years, suddenly had a big fallout ... as it was said that almost all ataris were suffering from bad hardware. The DMA chip in the Ataris, all except one ... were labelled "BAD DMA".


"SHOCK HORROR!" headline styling aside, are you referring to the early issue STE's perchance?



:coffe:

Yeb, that is the issue ... I've got two right now, one with good DMA, and one with BAD DMA. I did have some issues with it, a lot ... took a looong time to tune it up. I did this in software, on a MIPS running at 80MHz, and what I discovered is that if the handshake line, is held low too long ... bang! The maximum values, given by ATARI of 250ns ... don't add up. If they are our of specs with the DMA chip, the data will be garbled.

User avatar
CiH
Atari God
Atari God
Posts: 1134
Joined: Wed Feb 11, 2004 4:34 pm
Location: Middle Earth (Npton) UK
Contact:

Re: The BAD DMA Chip myth

Postby CiH » Tue Apr 17, 2012 2:18 pm

Looks like you went to a lot of trouble.

So in more general terms, is the fix something as simple as a software patch which can be applied to existing set-ups without hardware modification or addition, or is a combined hardware/firmware approach needed?
"Where teh feck is teh Hash key on this Mac?!"

User avatar
christos
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2427
Joined: Tue Apr 13, 2004 8:24 pm
Location: Greece
Contact:

Re: The BAD DMA Chip myth

Postby christos » Tue Apr 17, 2012 2:21 pm

I am more interested in the device. How does it look like, where does it work, can you make it for scsi so it can be used on falcon, and how much did it cost to make?
Felix qui potuit rerum cognoscere causas.
My Atari blog

STOT Email address: stot(NoSPAM)atari(DOT)org

User avatar
Cyprian
10 GOTO 10
10 GOTO 10
Posts: 1626
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: The BAD DMA Chip myth

Postby Cyprian » Tue Apr 17, 2012 2:29 pm

oehansen wrote:For the past 5 years, the Atari Community has apparently almost dissappeared, as far as I can see. The Atari having been alive for years, suddenly had a big fallout ... as it was said that almost all ataris were suffering from bad hardware. The DMA chip in the Ataris, all except one ... were labelled "BAD DMA".

It's time to shake this myth ... it's time for some "myth busting".

I made my own device, to read CF cards, and USB "thumb drives". I use a cheap PIC32MX795 based solution, called Max32. I've played around with, and have weeding my bugs out, and playing with this beast for a couple of months. All on my "BAD DMA" Atari. I use HDDRIVER, on the Atari. And have tried MINT, as well as normal atari. I have HDDRIVER boot off the C partition on the drive, and HDDRIVER now recognized the "Thumb drive" as more than 1GB, using group 1 SCSI commands. I even got myself an Atari with a good dma chip, for the sake of safety.

What I did discover, is that the DMA chip is enormously sensitive to timing. There are two different timings on the chip, one for COMMANDS. And one for DATA. You CANNOT use the same timing, for DRQ and IRQ signals. I did the timing in software, rather than with Hardware. I tuned in the signal return, until it worked ... on both the BAD and the GOOD chip Ataris.

So, for all you people out there, that have an Atari with a bad DMA chip. You don't have to throw them away ...



it isn't a myth, faulty DMA is proved issue. Here you can find well documented proof: http://joo.kie.sk/ultrasatan/steproblem/
Lynx / Jaugar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
SDrive / PAK68/3 / Lynx Multi Card / LDW Super 2000 / XCA12 / SkunkBoard / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Steem SSE / Aranym / Saint
http://260ste.appspot.com/

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Tue Apr 17, 2012 3:10 pm

Basically, I think Atari shot themselves in the foot on this one. I've been hammering on this DMA chip for months ... literally, and this is the original DMA chip.

First, and most basically ... the chip wouldn't work with Floppy either, if it really was faulty. I read on another forum, where someone pointed out that it was most likely fixable by replacing the 74*244 that was between the port and the DMA. Actually, I *DID* replace that one with a 74F244.

Here are some images of the device, I use. If someone notices that it is a Max32 with a network shield on top, yes I noticed that too. I actually intend to implement the "scsi to internet" protocol, as there is already a driver for it (I haven't started with that yet.). And in case someone notices that there is a PS/2 connector there, with an ATARI iKBD connector as well. Yeb, it sure is.

Otherwise, when interfacing with HDDRIVER, I noticed there is an option ICD extension. Which basically is, that they put an opcode 1F ahead of their extension. So, yes, scsi is the best option as the "Thumb drives" all talk scsi. So does a USB connected card reader.
You do not have the required permissions to view the files attached to this post.

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Tue Apr 17, 2012 3:54 pm

CiH wrote:Looks like you went to a lot of trouble.

So in more general terms, is the fix something as simple as a software patch which can be applied to existing set-ups without hardware modification or addition, or is a combined hardware/firmware approach needed?


I don't think the replacing of 74LS244 with a 74F244 was really needed. I did that, because I was beginning to think "maybe" it was a hardware issue, and it *REALLY* was faulty. The replacement didn't change a thing. I could always get the commands over, but not the data. Here is what I did, in software ... on the PIC32 at 80MHz, a loop *for(int i=0;i<100;i++)" is a 10us delay loop. However ...

I program a delay, between bytes ... the delay is done with a timer ... I have one signal, HANDSHAKE, and it is multiplexed between IRQ and DRQ (makes sense, right?). Now, entering here, I need to do this differently between DATA and COMMAND. Receiving or Transmitting DATA, it literally has to release the HANDSHAKE line about 30ns after ACK goes low in reply. In the following, I cannot even do a "for(int i=0;I<2;i++)" ... that is too long. A "for(int i=0;i<n && PMRD == HIGH;i++)". That is ALSO out of it. The reason is, that PIC32 is a MIPS processor, and it's timing can go pretty off 12.5ns per instruction, but not into the 100's of nanoseconds though. Which literally means, that for this to work, the HANDSHAKE line has to go HIGH, somewhere around 30ns after CS goes low (about 12.5ns per instruction as mT4ClearIntFlag is just a macro to set one register. In no way, can it wait the 250ns attributed by Atari.

Code: Select all

            case DELAY:
                      if( debugging == 2 )
                          debugPrint("Handshake at %d ticks\n", timeout_value);
                      ahdiInfo.timer = TIMEOUT;
                      timeout_value = TICK_RATE * 3;
                      if( ahdiInfo.state == DATA_SND )
                      {
                          HANDSHAKE = LOW;
                          if( PMRD == HIGH )
                            if( PMRD == HIGH )
                              if( PMRD == HIGH )
                                if( PMRD == HIGH )
                                  if( PMRD == HIGH )
                                    if( PMRD == HIGH )
                                      if( PMRD == HIGH )
                                        if( PMRD == HIGH )
                                          if( PMRD == HIGH )
                                            if( PMRD == HIGH )
                                              if( PMRD == HIGH )
                                                 ahdiInfo.errorCode = 1;
                          mT4ClearIntFlag();
                          HANDSHAKE = HIGH;


However, if its a command, the HANDSHAKE can be stretched further. I put up the HANDSHAKE signal, only when the byte is actually received (at the end of CS).

Code: Select all

void __ISR(_PMP_VECTOR, ipl3) pmpHandlerISR(void)
{
    int data;
   
    HANDSHAKE = HIGH;
    ahdiInfo.timer = INVALID;
    if( PMSTAT & 0x8000 )
    {
        switch( ahdiInfo.state )
        {


User avatar
alexh
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2723
Joined: Wed Oct 20, 2004 1:52 pm
Location: UK - Oxford
Contact:

Re: The BAD DMA Chip myth

Postby alexh » Tue Apr 17, 2012 4:10 pm

With the timing information you've gathered and the software work around you've discovered, is it possible to feed back to the author of HDDriver so they can produce a driver which works with all revisions of DMA chip without affecting performance?

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Tue Apr 17, 2012 4:25 pm

alexh wrote:With the timing information you've gathered and the software work around you've discovered, is it possible to feed back to the author of HDDriver so they can produce a driver which works with all revisions of DMA chip without affecting performance?


This is not something that could be done inside the Atari, but something that would have to be done inside "SatanDisc" for example.

User avatar
rocket-dog
Atari Super Hero
Atari Super Hero
Posts: 779
Joined: Wed Mar 05, 2008 3:13 pm
Location: Great Britain - The Kingdom of Mercia

Re: The BAD DMA Chip myth

Postby rocket-dog » Tue Apr 17, 2012 9:05 pm

I have a Mega with a duff DMA. Neither my US or Syquest drive works. And nor does my external 3.5 drive which I believe is controlled by the same chip. The onboard floppy works just fine.

User avatar
wongck
Ultimate Atarian
Ultimate Atarian
Posts: 12672
Joined: Sat May 03, 2008 2:09 pm
Location: Far East
Contact:

Re: The BAD DMA Chip myth

Postby wongck » Tue Apr 17, 2012 11:15 pm

oehansen wrote:Here are some images of the device, I use. If someone notices that it is a Max32 with a network shield on top, yes I noticed that too. I actually intend to implement the "scsi to internet" protocol, as there is already a driver for it (I haven't started with that yet.). And in case someone notices that there is a PS/2 connector there, with an ATARI iKBD connector as well. Yeb, it sure is.


8O OMG... if you pull this one off, it will be another one of those upgrades that ever ST user should get. [smilie=greencolorz4_pdt_12.gif]
My Stuff: FB/Falcon CT63 CTPCI ATI RTL8139 USB 512MB 30GB HDD CF HxC_SD/ TT030 68882 4+32MB 520MB Nova/ 520STFM 4MB Tos206 SCSI
Shared SCSI Bus:ScsiLink ethernet, 9GB HDD,SD-reader @ http://phsw.atari.org
My Atari stuff for sale - click here for list

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Wed Apr 18, 2012 7:39 am

rocket-dog wrote:I have a Mega with a duff DMA. Neither my US or Syquest drive works. And nor does my external 3.5 drive which I believe is controlled by the same chip. The onboard floppy works just fine.


IFF the DMA chip was "really" BAD, then the on-board one shouldn't work either.

In your case, I'd look at these threads.

http://www.google.com/url?sa=t&rct=j&q= ... niZkw4iRyA

But it's not "merely" the signal levels, as someone pointed out. The onboard PMP of the PIC32MX is TTL compatible, however. The current it uses, is low ... just as the SatanDisc is.

I have a theory, and it goes like this. DMA on my Atari *DID* change, between I first started and now ... there was a characteristical change in between. Functionally, going from "easy" to "hard" to implement. According to Atari's specifications, the ACK line will go low no later than 60ns after DRQ goes low. This one fits well, but it also says that the ACK line can stay low for about 250ns, this part doesn't fit. The DRQ line *MUST* go high *BEFORE* ACK does, for the DMA transfer to work, and this is merely some 30ns to 60ns or so. It's about 3 "sequential" instructions on a MIPS32 at 80MHz. But most devices (if not all) are counting on a 250ns window, to set the DRQ high before the ACK line goes high. If the DRQ goes high close to the time ACK goes high ... you get garbage data. Now, if you have chips, like the 74LS244 and 74LS255 which are driven by very low current, and these have to drive a "fat" copper line at relatively long distances (compared to the current available), then you may get a similar condition. The buffer chips, can cause a faulty rise/fall times of the signals, and bring the ACK/DRQ lines too close to each other. However notice, that the CS/IRQ lines do *NOT* suffer this situation.

The fault inside the DMA chip has to do with that the ACK line timeout, cannot be stretched the full cycle time of 250ns. Something inside the chip, will "break", perhaps a capacitor to a monostable? causing the ACK line to stay low, no longer than a 30-60ns window.

But, it doesn't mean that the DMA chip is unusable ... especially since the onboard floppy seems to work properly, does it?

What was the cost of the device, I made?

I love electronics, so I got myself a Max32 for the fun of it, and a Network Shield for it. Basically because I "intend" to have it do all the work for the Atari, Internet, USB, SCSI. The max32 is around $50, and the network shield similar. A protoshield board, is abot $6.50, and then a GAL chip, which is around $2 bucks. And my SanDisc 8GB that I am using with the Atari, was about $14.

All in all, around $120.

User avatar
alexh
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2723
Joined: Wed Oct 20, 2004 1:52 pm
Location: UK - Oxford
Contact:

Re: The BAD DMA Chip myth

Postby alexh » Wed Apr 18, 2012 12:51 pm

oehansen wrote:This is not something that could be done inside the Atari, but something that would have to be done inside "SatanDisc" for example.

AFAIK the UltraSatan project is opensource?

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Wed Apr 18, 2012 1:15 pm

alexh wrote:AFAIK the UltraSatan project is opensource?


I have a blackfin in my drawer, so I plan on building one as well ... as soon as I get a proper setup for etching boards. Currently I don't have that setup ... no laser printer, and no need to get one. So, I'll have to do a photo positive etching ... maybe in a month or two, I'll have that setup.

But currently, I'm having way too much fun with my good'ol'atari. I'm on a nostalgia trip here ...

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Wed May 02, 2012 8:38 am

alexh wrote:
oehansen wrote:This is not something that could be done inside the Atari, but something that would have to be done inside "SatanDisc" for example.

AFAIK the UltraSatan project is opensource?


What I've seen about "OpenSource", and the word "Open", does not mean it's open. It's like "Free" ... in reality, neither "Speech" nor "Beer" is free.

This will always be one of those projects, where you'll have to do it alone. *Maybe* I'll do the board itself, but I am not going to produce it and sell it. The reason is simple, I've been debating with myself wether I should buy the FireBee or not. It's a nice piece of hardware, but just like with the Bad DMA chip, I think it's basically a dead end. Another "will fade into nothingness", and be one of those, so many now, "An atari only for me-and-my-pals" type. I don't think the "GOATARI" version of Commodore 128 "Go64" is a good idea. And I don't want to have to run software in emulators, or have to have a different screen, keyboard and basically *everything* to run my atari stuff. I feel it lacks vision, and understanding of basic computation, history of computation, where we came from and where we're headed. MiNT is a POSIX system, or strives to be ... but apparently, that doesn't mean a lot to a lot of people. All I hear is "I'm a professional computer genius", and I have a tendancy to answer "No, you're not ... your a worker, who has zero vision and the only reason you have your Job, is because you are NOT Mr. Gates, Jobs, Wozniak, Kendrall, Kernighan or Richie. And no ambition to strive to be. You're a guy with no understanding of these issues, and no vision of where to take it, so your superiors do not see you as a threat, you are not a guy who may take this work further and make his own stuff out of it. You're not Mr. Gates, whose gonna rip IBM off, nor are you going to take Kendrall's CP/M and make it into QDOS.". But, that's basically an insult, so most of the time I revert from saying it, because we can always get philosophical about it, and discuss whether it's actually wise for the salmon to swim against the current. I'm too old to get philosophical.

But I'll share this, with anyone who wants it. This needs a MAX32, and is designed as a MAX32 Shield. It's simpler that way, in ways of programming and accessing it. The 555 is used to create a monostable, that can have a varying timeout. Using the monostable 1.1*R*C, it should be easy to get a good timeout. But, beware ... this is the heart of the problem with the Bad DMA chip, because in reality you can't do this. The timeout for the DMA, is to short ... so J1 is put in, where it can be easily selected, what signal is used for the handshaking. In the end, you'll need two ... especially if you have a BAD DMA chip, which means software is the only solution. To adjust the resistor, is not that easy ... but if you can do it, it's the perfect solution though. I skipped this part, personally, and was satisfied with the software solution. All I wanted was a working hard drive, don't care much about the rest. There is also one wire missing in the drawing, From RA4 to I6 of the GAL. The programming of the GAL should be obvious.

IRQ = H or DIR
DRQ = I6 or !DIR
PMWR = R/W or PMCS1
PMRD = !R/W or PMCS1
PMCS1 = (CS or DIR) and (ACK or !DIR)
SCL1 = A1 or CS

The hard drive isn't the fastest, it can go 133 Kb/s (Kilo bytes), but not much more than that. But that is the USB Host controller, that defines the limitations.

The MAX32 Board, can be easily replaced with a PIC32, but that requires that its Host interface be connected. A PIC32 is cheap, and easily come by ... you'll end up with a board for less than $50. I've tried the board with HDDRIVER and it works very well. But it's up to you to "translate" the SCSI(6) commands to SCSI(10) commands in many cases, as some of the larger Thumb drives, do not support the READ(6)/WRITE(6) commands, even if the Standard says they should.

Electrical design.png
You do not have the required permissions to view the files attached to this post.

oehansen
Captain Atari
Captain Atari
Posts: 291
Joined: Tue Apr 17, 2012 12:05 pm

Re: The BAD DMA Chip myth

Postby oehansen » Wed May 02, 2012 9:21 am

The signal levels of this cirquit draws very little current, if it's done by putting a PIC32 there instead of a MAX32, and by removing the diodes. You could basically run it off the Atari power, through the pullup resistors. This is another possibility, the only problem here is that when resetting the Atari resets faster than the stock MAX32. You'd have to dump the bootstrap program on the PIC32, to ensure that reset would result in the PIC starting first. I ignore the reset lines, and simply turn on the unit prior to the atari. Again, I got a working unit ... the rest is optional.

joska
Hardware Guru
Hardware Guru
Posts: 4343
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: The BAD DMA Chip myth

Postby joska » Wed May 02, 2012 9:44 am

oehansen wrote:MiNT is a POSIX system, or strives to be ...


No, it's not. Please take a look at the POSIX requirements and you'll see why.
Jo Even

VanillaMiNT - Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

joska
Hardware Guru
Hardware Guru
Posts: 4343
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: The BAD DMA Chip myth

Postby joska » Wed May 02, 2012 10:21 am

oehansen wrote:Yeb, that is the issue ... I've got two right now, one with good DMA, and one with BAD DMA. I did have some issues with it, a lot ... took a looong time to tune it up. I did this in software, on a MIPS running at 80MHz, and what I discovered is that if the handshake line, is held low too long ... bang! The maximum values, given by ATARI of 250ns ... don't add up. If they are our of specs with the DMA chip, the data will be garbled.


I'm not sure what you mean by "myth", as you say yourself that there's a "bad" DMA-chip. But that aside, your device looks very interesting. Do you intend to share the code for the PIC?
Jo Even

VanillaMiNT - Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

User avatar
alexh
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2723
Joined: Wed Oct 20, 2004 1:52 pm
Location: UK - Oxford
Contact:

Re: The BAD DMA Chip myth

Postby alexh » Wed May 02, 2012 11:02 am

oehansen wrote:
alexh wrote:
oehansen wrote:This is not something that could be done inside the Atari, but something that would have to be done inside "SatanDisc" for example.

AFAIK the UltraSatan project is opensource?


What I've seen about "OpenSource", and the word "Open", does not mean it's open. It's like "Free" ... in reality, neither "Speech" nor "Beer" is free.

As competant as I'm sure you are. And as interesting as your discovery is. You do talk a load of bollox! :cheers:

http://joo.kie.sk/ultrasatan/download/

The PCB design is open source and free.
The firmware is open source and free.

If as you say (and I don't doubt it) that the problems of the "bad DMA" chip can be worked around. And if they can be worked around in firmware (or small mods to the PCB) with UltraSatan... then surely this is the perfect starting point? Plus it already has users.

User avatar
Official Ninja
Captain Atari
Captain Atari
Posts: 155
Joined: Tue Mar 29, 2011 1:37 pm
Location: New Jersey, USA

Re: The BAD DMA Chip myth

Postby Official Ninja » Wed May 02, 2012 4:05 pm

I have a 1040STe US with TOS 1.62
Does the 1.62 TOS mean that there is a good chance I have one of these bad DMA chips?

joska
Hardware Guru
Hardware Guru
Posts: 4343
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: The BAD DMA Chip myth

Postby joska » Wed May 02, 2012 6:42 pm

The "myth" has it that only early STE's had this problem. They had TOS 1.06 fitted, so if your STE was originally fitted with 1.62 I think you should be safe.

http://members.optusnet.com.au/~colbee1/DMAFAULT.HTM
Jo Even

VanillaMiNT - Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

Shredder11
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2434
Joined: Mon Nov 26, 2007 2:23 pm
Location: West Yorkshire, United Kingdom (England)
Contact:

Re: The BAD DMA Chip myth

Postby Shredder11 » Wed May 02, 2012 9:32 pm

My first 1040STE came with TOS 1.06 but had the good DMA chip.


Social Media

     

Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 6 guests