Improving 800XL SIO transfer rate.

[+mytek]

 

 

You are absolutely right! My eye-sight is so poor I never noticed that even on the blown-up photo!

 

I wonder what it was stuck between the 'AUDIOIN' and 'GND' for? This particular machine is one that, long ago used to belong to the leader of an Atari cracking group down in Portsmouth. I can only assume he must have though it would help in some arcane way - perhaps to improve 'CLOAD' reliability? Whatever the case it is gone now and SIO runs like a machine-gun between the "SIO2SD" and the machine!

 

It still could be a capacitor, because I have seen some in the past that are packaged just like a glass diode, but likely not. Glad to hear that your SIO is blasting away.

 

I just can't help but thinking how sweet it would have been if Atari had implemented high-speed from the get go. That would have made the C64 disk access look like a snail by comparison. And from what I understand, the creator of the SIO interface went on to design USB. If higher speed had been stock, the SIO would have faired pretty well against USB 1.0 standard.

 

-Michael

[morelenmir]

 

It still could be a capacitor, because I have seen some in the past that are packaged just like a glass diode, but likely not. Glad to hear that your SIO is blasting away.

 

I just can't help but thinking how sweet it would have been if Atari had implemented high-speed from the get go. That would have made the C64 disk access look like a snail by comparison. And from what I understand, the creator of the SIO interface went on to design USB. If higher speed had been stock, the SIO would have faired pretty well against USB 1.0 standard.

 

-Michael

 

I have read that as well. It is fascinating to note how many - small though they may have seemed at the time - concepts that were built in to the fabric of the A8 in the late seventies and early eighties went on to be huge mainstays of the PC ecosystem more than fifteen years later. Just the PBI drivers thing - pre-dating 'plug-and-play' on the PCI bus by how long? At least ten years?

 

They were so advanced and then... I don't know... It makes me so hugely sad to think what happened when Warner sold out to Tramiel... I know there are some people who laud the dude, but I cannot forgive him for what happened to the XLD's and the rest that were so close to being released. If only Warner had give them another 18 months... Ahh, well. It is what it is.

[+mytek]

 

I have read that as well. It is fascinating to note how many - small though they may have seemed at the time - concepts that were built in to the fabric of the A8 in the late seventies and early eighties went on to be huge mainstays of the PC ecosystem more than fifteen years later. Just the PBI drivers thing - pre-dating 'plug-and-play' on the PCI bus by how long? At least ten years?

 

They were so advanced and then... I don't know... It makes me so hugely sad to think what happened when Warner sold out to Tramiel... I know there are some people who laud the dude, but I cannot forgive him for what happened to the XLD's and the rest that were so close to being released. If only Warner had give them another 18 months... Ahh, well. It is what it is.

 

Speaking of XLD's; I wanted a 1450XLD sooooo bad when I first heard about it. It had everything I could have asked for (great keyboard, sleek design, and a parallel interfaced double-sided double density drive built-in). Well a built in IDE hard drive would have also been nice. As for the built-in modem, kinda nice, but it would have soon been obsoleted as in too slow (this would have been better served by having it be a plug-in module, perhaps having the floppy storage area being used for this purpose, so that it could be easily upgraded or as a way to use other different types of modules).

 

To say the least, I was greatly disappointed when I couldn't buy one

 

-Michael

[morelenmir]

 

Speaking of XLD's; I wanted a 1450XLD sooooo bad when I first heard about it. It had everything I could have asked for (great keyboard, sleek design, and a parallel interfaced double-sided double density drive built-in). Well a built in IDE hard drive would have also been nice. As for the built-in modem, kinda nice, but it would have soon been obsoleted as in too slow (this would have been better served by having it be a plug-in module, perhaps having the floppy storage area being used for this purpose, so that it could be easily upgraded or as a way to use other different types of modules).

 

To say the least, I was greatly disappointed when I couldn't buy one

 

-Michael

 

Yup... Incredibly frustrating...

 

I got a smile - among many - when watching 'Kung Fury' the other week and seeing how much retro gear, including Atari stuff they used for set decoration and props. I am sure there was a 1400XL case in there at one point!

[+DrVenkman]

I've got my original early '84 Atari catalog that prominently shows the 1450XLD and 1400XL - I picked it up at a computer store and lusted after it for months until the Inc.-to-Corp. business. Alas.

 

At any rate, morelenmir, do you have photos of the mod you performed to increase SIO performance? Once I've got the U1MB installed in my 1200XL, I'll give the SIO mod a shot in one of my 800XL machines.

[morelenmir]

I've got my original early '84 Atari catalog that prominently shows the 1450XLD and 1400XL - I picked it up at a computer store and lusted after it for months until the Inc.-to-Corp. business. Alas.

 

At any rate, morelenmir, do you have photos of the mod you performed to increase SIO performance? Once I've got the U1MB installed in my 1200XL, I'll give the SIO mod a shot in one of my 800XL machines.

 

Sure thing!!!

 

I'll go take some snaps after 'av 'ad me supper - as we say North of the border.

 

I have one of those catalogues as well - I got it with my 800XL that was itself a remnanted bundle with a 1010 from Boots the Chemist for £120 in October 1985. How I pored over the 1400's and that light-scribe pen thing... Oh I wanted one of each so horrifically badly.

[morelenmir]

I've got my original early '84 Atari catalog that prominently shows the 1450XLD and 1400XL - I picked it up at a computer store and lusted after it for months until the Inc.-to-Corp. business. Alas.

 

At any rate, morelenmir, do you have photos of the mod you performed to increase SIO performance? Once I've got the U1MB installed in my 1200XL, I'll give the SIO mod a shot in one of my 800XL machines.

 

Here you go! It is just about the simplest hardware mod you can find. Just yank out the two capacitors which are connected to the DATAIN and DATAOUT lines - 'C76' and 'C77' - and then solder a 4700ohm resister between Pin 10 and Pin 5 of the SIO socket. It takes genuinely less than 10 minutes, and if you're not trying to be fancy with heat-shrink sleeving then probably under five!

 

Obviously you won't see any sped improvement unless you have an SIO device that is itself capable of better throughput than the standard, crippled SIO specification. Those that can make use of it really make use of it though!

 

As a matter of interest this is not my primary motherboard as when the "Ultimate1MB" is bolted in place it totally obscures the area of interest. However, this machine also needed doing so getting these photos gave me an excuse to sit down and do it. The mod is identical and I will now get super-fast SIO if/when I bring this 'REV D' unit back to life as well!

[GoodByteXL]

Here you go! It is just about the simplest hardware mod you can find. Just yank out the two capacitors which are connected to the DATAIN and DATAOUT lines - 'C76' and 'C77' - and then solder a 4700ohm resister between Pin 10 and Pin 5 of the SIO socket. It takes genuinely less than 10 minutes, and if you're not trying to be fancy with heat-shrink sleeving then probably under five!

Read and heard this so often ...

To me it seems that these fixes are only needed when improper loads on the SIO happen caused by devices like SIO2SD or S:Drive etc..

I've never ever needed to do this mod on any XL/XE/XEGS to run high speed SIO drives down to Pokey Divisor 0.

 

All high speed floppy disk drives ever available were tested during the decades and worked fine, also SIO2USB(ABBUC). SIO2PC(USB) based on original FTDI breakout boards works perfect. Only SIO2SD and S:Drive had problems and that's why they were given away shortly after having them aquired. I think if these fixes are needed, there is something wrong in the system.

 

But that's just a personal remark. If it works for you, fine :-).

[Rybags]

(diode between Gnd / AudioIn)

 

Maybe it helps clear up the crosstalk that comes into audio in when loading from tape, though that would probably make zero difference to load reliability and no difference for other devices.

Edited July 28, 2015 by Rybags

[Joey Z]

(diode between Gnd / AudioIn)

 

Maybe it helps clear up the crosstalk that comes into audio in when loading from tape, though that would probably make zero difference to load reliability and no difference for other devices.

Are we sure it's even a diode? atari used capacitors that looked just like that on many of their machines. I don't think a diode would make sense here. It would either ground the audio out, or it would be placed in the other way and do nothing.

[morelenmir]

Are we sure it's even a diode? atari used capacitors that looked just like that on many of their machines. I don't think a diode would make sense here. It would either ground the audio out, or it would be placed in the other way and do nothing.

 

It is certainly weird... That diode symbol on the component is quite persuasive, but who knows. I would have tested it, but its gone now. Ah well.

 

In regards this mod's utility. It seems to work for me - at least for the most part with one notable exception. Hope the pictures help!

[+DrVenkman]

I finally got around to doing this - and can detect an appreciable difference in the transfer rate even before adding the resistor. On inspecting the reverse of my SIO socket in order to do so I found a 'bodge' component already in place, but not I think a resister. See attached photo.

 

Neither of my other 800XL's have these added, so they were possibly already removed by their previous owners or this one received special attention for some reason. Is it safe to get rid of ot do you think?

 

As a matter of interest how critical is the resister rating? Does it have to be 4700 ohms or is that just what was convenient at the time? Annoyingly I don't have that precise value.

Long story, but I was inside my spare 800XL today and note that it's a Rec C board (my main unit is a Rev A). It too has a "bodge" component like yours in the bottom of the board between two SIO pins, though it looks much more like a resistor. I didn't photograph it, as I had left my phone in the other room and. I was busy, but I immediately though of this post. The board also has two factory-installed jumpers on the top of the board running along the right side. I'll eventually open it back up again and take pics. Edited August 1, 2015 by DrVenkman

[1050]

I've had all I can stands and I can'ts stands me no more !!!

Old school clips starring bodge the cap AKA the nefarious masquerader

 

 

 

 

 

 

 

 

 

 

 

credits first - what's this on?

 

 

 

 

 

 

 

 

 

 

 

Uh-huh, bet you can find this one on yours too, but it might not look like bodge does here. You see C107 hopefully? That's a capacitor because the silkscreen say so. Look closer - you see the 104 on it? That's shorthand for 10 followed by four zeros, 100,000 picofarads or better known as .1 mfd - a standard value for a bypass capacitor also known as a decoupling capacitor.

Selecting Decoupling Capacitors for Atmel’s PLDs
http://www.atmel.com/images/doc0484.pdf

Can you not see the surface mount capacitor that Atari for whatever insane reason they might have had to do so, wound up putting dear old bodge inside this prison of glass? You are aware that glass encapsulated diodes are marked with a solid black band near their cathode end all the way around so you can't miss it. Do you see that anywhere around, good. No that's right because it isn't a diode at all. It's a cap.

 

 

 

 

 

 

 

 

 

 

 

Oh my goodness, that seems vaguely familiar does it not? But wait, it's got five bands - huh.

 

 

 

 

 

 

 

 

The reveal

 

 

 

 

 

 

 

 

More five banded armadillos? Not what the silkscreen says.

This is just south of the SIO jack on the component side - I didn't realize myself they were all caps there. all the way down to the bottom of the joystick port J5 also on the silkscreen. Brown is one, black is zero, red is 2 zeros for 1,000 picofarads or .001mfd - I cheated and looked at my schematic. Sue me. I actually do know the number method, these particular caps are often baby blue with 102 on their sides, look long enough and you might see one in this row of caps showing everything. On this same board old bodge is to the left of every single ram chip, not one of them shows me much of anything clearly except for one and all it says 50 V.

As Kyle22 simply said a long time ago too - it's a capacitor.

[+DrVenkman]

 

0801150044a.jpg

 

Oh my goodness, that seems vaguely familiar does it not? But wait, it's got five bands - huh.

 

(SNIPPED)

 

As Kyle22 simply said a long time ago too - it's a capacitor.

 

 

Pardon the shitty camera on my iPad - my phone was in the other room while I was messing about last night - looks like brown, black, red, silver, silver:

 

 

This board also has a couple small components and long jumper wires running along the length of the right side of the board up to the SIO area. I didn't take photos because, like I said, my phone was elsewhere and I was busy.

 

But to ME the interesting question is: why are they there on this Rev C board and NOT there on my Rev A board? What was Atari getting at with these additions to the base design?

Edited August 1, 2015 by DrVenkman

[morelenmir]

Its a fascinating topic chaps! Its also not on my REV D...

[1050]

Can't know the why of these additions from Atari's side for sure, but we can guess. They often did these mods even on the later XE runs you'll find jumpers on some and not on others with no other difference. The board I pictured also has the two hot wire jumpers up to the SIO area back feeding 5+ power to the BASIC rom and SIO jack. Someone in final assembly plant with some authority over the minions wanted this done and it was done.

 

BUT a huge safety issue is involved here - that direct jumper to the SIO jack will connect power buss directly to the +5 READY line of the SIO jack and that's none too smart. It's is supposed to be isolated by L10 inductor which is the only safety valve used. Other jumper to BASIC rom is fine and safe, I will remove the jumper to bottom end of L10 on this one, I suggest you all do the same. Had I not looked closer because of this follow up, I wouldn't have known they could be so stupid at the final assembly plant, so thanks guys.

 

All this time I assumed they had some notice from Atari that they were operating on to install these kinds of factory jumpers - well that idea is now in the round file - this guy was an IDIOT and this is a really dumb thing to do. And it puts these factory jumpers in a completely different category for me. Wannabe ace engineer can't make it anywhere else and he feels the need to re-engineer things under his POWER. Power tripping tool head in charge - look out. These kind of guys are the ones you don't want anywhere near the launch button of a nuclear weapon, huh?

 

If two computers are connected via the SIO cable, as some are wanting to do, this one would be powering the other one and that's NOT what the READY line in the SIO cable is for - the SIO cable can't take the current for just one reason to not have it this way, so the cable is at risk for burning the wire out of it. Once that happens, no drive fed thru that cable will operate ever again.

 

There is no defect that the SIO jumper cures, it's just not kosher and I won't have it under my roof. Only reason the world doesn't stop spinning because of this is because this READY line is only used to let drives know the computer is on and READY. No other use for it is the only reason we don't have a very BIG problem and issue with these jumpers - I still won't have it.

[_The Doctor__]

um, having the sio with full complement of power is a thing of beauty (recalling people hacked off and removed power from their 800's), what was really wrong is not being isolated using a diode and a fuse with a current limiting resistor... so If you are a purist I would put back what was always intended but fix the safety issue with a simple diode and fuse... no muss no fuss, and one less power block needed to run the alien voice group synthesizer box and other devices designed to use the different powers intended to be supplied by SIO..... yes I also put the power mod to return the 1200XL sio back to being useful. The modification to add power to 'stabilize' the sio signal is fine so long as a diode, fuse and possible current limiting resistor is in place and kept in proper protective order. It does scope out and if it's done right, perfectly safe. Don't skimp on such an inexpensive fix, really the cost and time for a diode resistor and fuse is well worth it... I mean pocket change mere coins... you will spend more on a beverage even!