EmpireAndrew wrote:So you always put 2.2k's in to replace the red and black ones?
Basically yes. Though one is also replaced with 1.2k if its not already (P100).
EmpireAndrew wrote:But I take it that in itself doesn't necessarily fix the issue or you wouldn't be adding a resistor pack to the -38?
The resistor on the -38 is my first find of issue. The motherboard resistors are my second find. Though nobody has tried it to solve DMA problems yes. I never had any real issues with the DMA so I have no idea if it will solve the DMA issues are not.
EmpireAndrew wrote:And I agree, it would be good to know what is different about the 001 that (mostly?) fixes people's problems.
People are always so quick to change to the 001A without trying other things first. So I just don't know if there is any actual faulty -38s or not.
EmpireAndrew wrote:Atari obviously switched to it for a reason, but maybe it simply tolerates some other underlying issue with the board design better than the 38?.
EXACTLY!!! finally someone talks some sense on here.
EmpireAndrew wrote:Frankly, I just want it working. I spent a lot to get this STE as I wanted as close to perfect as I could get so I could just go ahead and enjoy the machine. So it's frustrating to have to perform surgery on it.
You don't even need to change the resistor packs to test it out, just solder them on the back of the ones on the motherboard.
EmpireAndrew wrote:But I'd like to know why whatever fix I do actually fixes it. Ridiculous that 25 years on we don't have an answer.
Yes, I've been TRYING to explain these DMA issues for years and pretty much given up. This will definatly be the last DMA thread I will replyl to, as I am just repeating myself over and over, just doing nothing but wasting my time.
I'm trying to help people finally solve DMA issues, but everyones brain washed into thinking every -38 on a STE is faulty. Simply not true. I've only once seen a faulty -38. As to if it was produced that way or just failed over time is unknown.
The "rumor" is *some* very early STE had bad DMA chips. Apparently very few. But pretty much everyone who has a -38 thinks it is faulty and HAS to replace it.
EmpireAndrew wrote:But there will be a limit on how much investigation I am prepared to do, in any case I don't own a scope or anything so there isn't much I can do anyway.
Yes I can appreciate that. You dont need a scope, just the resistor packs soldering on. If it solves your problem then its a huge step into understanding the real cause of the DMA issues people are having, if it doesn't solve it, I will keep looking. Currently I can't do anything more as nobody will help.
The last guy who had DMA issues sold his machine to troed on here, who said the DMA seems to work fine. So Is there DMA issues or isn't there...
EmpireAndrew wrote:I think I'll start with the resistor pack on the 38, after all the only other option is to replace it anyway so I'm still faffing with it regardless.If that doesn't work I'll probably swap it out tbh.
If you put the resistor pack on the -38 and it doesn't help, then you need to update the motherboard ones. If that doesn't work, then you really have no choice but to change the DMA.
EmpireAndrew wrote:And if that doesn't work then I'll put a resistor pack on the 001.
I've not done much with the 001A, its possible it needs pull downs and pull ups on it, so its not so simple, I've not spent any time looking for 001 fixes.
EmpireAndrew wrote:Only then will I mess with the rest of the board (i.e. the other resistor packs).
The less I mess with the board itself the better. Less chance of screwing up.
If you just tag the resistor packs on the bottom of the motherboard, then as long as you actually have them soldered and not shorted, then in theory your motherboard should match mine with the -38. Don't forget the red one towards the front of the motherboard should be 1.2K (P100).