Microbotics M1230XA problem, possibly faulty?

[gurrag]

A sharp tip on the soldering iron, might have hit the slot a few times, but no big problem, since its just plastic.

[Amiga1992]

I see, I guess I need to work on my steady hand! :P
If I do this I should document it here, since finding the cap legs seems already quite a task on its own.

[Zetr0]

@Thread

This is a case of ockham's razor - the principle (attributed to William of Occam) that in explaining a thing no more assumptions should be made than are necessary.

Jumping onto capacitors / etc is to be honest - unless seen leaking - is grabbing at straws.

The only way to discover similarities that may allude to a conclusion you need to know and have tested identical kit (including software setup and hardware *memory / CPU mask*) - where the instance of failure is identical then you would have a possible source to the problem.

randomly jumping on capacitors, without some level of evidence to suggest it is a bit crazy.

[Amiga1992]

We have the exact same WHDLoad crash error, how is that not identical?
We determined capacitors are not at fault. His are new, mine aren't. S there is no " crazy" here anymore.

As I said: either Wepl gets my accelerator, if he cares, and checks it out, or we will not get a solution to this.

[BarryB]

I solved all my M1230XA problems by replacing it with a Blizzard 030!!

[ReadOnlyCat]

Originally Posted by Akira

We have the exact same WHDLoad crash error, how is that not identical?
We determined capacitors are not at fault. His are new, mine aren't. S there is no " crazy" here anymore.

As I said: either Wepl gets my accelerator, if he cares, and checks it out, or we will not get a solution to this.

I started to setup my 1200 today in the hope of reproducing the issue but I ran into some weird issues with two monitors suddenly stopping to work about 10 minutes after being plugged to the 1200...
Sniffing the kitten above Denise revealed a bit of a cat pee so I suspect my 1200's caps are leaking so I will have to fix that before I can give it a try.

Nevertheless, do not lose hope, I will do it.

[glenn]

Originally Posted by Akira

Yeah, because the probabilities of this happening are how high?

Fairly high yes. How many different cards have YOU tested it on ?

Just because you think your card works fine because you think you tested it fully eons ago doesn't make it correct. If you test the games now thoroughly, specifically the ones mentioned here with problems, with the latest version of WHDLoad, and they work, I'd be willing to take your suggestion. But as it is now, i's just speculation based on memory and positive bias toward your card.

I have a lot of Amigas with different accelerators, I have used whdload on most of those, there is no such thing as "everything work out of the box" on ANY card when it comes to whdload, but it seem to be a fact that the blizzard 1230 IV is the card that works best with whdload , I would guess that this because it is the most tested configuration (probably used by developers too).

(Regarding the "positive bias toward your card", It's not that I have ONE Amiga with ONE accelerator card, I have many Amigas with many cards of different makes and models.. the MBX1230 is just one in the crowd.)

Please bring facts to the thread (like "I unpacked my Amiga, tested it and I don't get these issues") if you wanna help.

Maybe you should ask your mom to teach you some manners instead ?

It all started with YOU saying "WHDLoad isn't compatible with this card." , and I responded to that since I used exact that configuration with whdload many many hours (after tweaking some switches), so it was simply wrong - and I pointed it out.

I also used it with other systems, and had to tweak switches to get it to work fine on those too.

If you don't like what I write, just ignore it, this is a discussion forum, where people discuss things, not everything in the thread is directed against you, and you do not have to respond to everything. You didn't even START the thread ffs..

[Zetr0]

Originally Posted by Akira

We have the exact same WHDLoad crash error, how is that not identical?

You have one instance, this at its very best is anecdotal (i.e. you need more data and testing to ensure how - when and why )

Originally Posted by Akira

We determined capacitors are not at fault. His are new, mine aren't. S there is no " crazy" here anymore.

Who is we!? your the one whom suggested this in the first place - I am commenting that jumping on a percieved solution with next to no information as to why a fault exists is just insane and will only leave you with less hair when that doesn't work.

fault finding is procedural and more often than not is slow levels of increment that will find the problem and the solution.

Originally Posted by Akira

As I said: either Wepl gets my accelerator, if he cares, and checks it out, or we will not get a solution to this.

You do like absolutes my friend, so your ingenious method of solving the problem is to shaft someone else with it?

Please show some restraint, and do your own studies - test the adaptor fully.

Start with the things you can test -
OS Wb31 - 3.9
Disk Format (FFS/SFS/PFS)
Memory, CPU's / with / without FPU's etc.
Card Configuration

build truth table - share your results and test builds with others and let them test them as well. It will be the commonalities that will determine if the card is at fault or at fault with a certain set of set ups.

Expecting someone to do this work for you (for gratis) is a tad absurd since there is quite a lot to it... if you show some effort in diagnosing others will help.

Start with a basic FFS and OS (nothing fancy)
Use the latest WHDLoad
Smallest Amount of RAM you have
Card in default configuration

TEST WHDLoad Game

if it fails, Test again, but change FFS to SFS and then to PFS - if it still fails go back to FFS and change the WHDLoad binary to an earlier one and repeat. if that fails then go to an earlier revision of WHDLoad again and repeat.

if that fails, you start again with latest WHDLoad binary, FFS and this time change memory (note the speed of RAM here -6,-7 or -8) and size

and repeat.

I think at this point you get the idea - nothing good comes easy

[Amiga1992]

Shaft someone with it? How am I shafting someone? I don't understand your issue here man, I said I will gladly lend my accelerator to Wepl for research for as long as he would need it, if he cares at all. I don't want to bother him with this. I think I am truly helping by offering to send out my device to Germany with all risk involved.

I think you need to show some restraint, you are being unnecessarily aggressive with me when I haven't done anything to you or anyone. All the things you have asked me to "test" I already did. Nobody told you to care about this problem really, so if you don't want to read back on what I have posted before, that's fine, but don't come at me saying that I *expect* anyone to do anything, when I said no such thing. I just don't want to do what I already did twice. If you don't want to "analyze" the problem, don't. it's a free board. I did not expect you to "work" for me "for free".

(also I know I suggested capacitor problems, but I am not a stubborn fuck and I took it back after hearing from gurrag).

GLenn, all I said was that your test of back then doens't apply with newest WHDLoad versions. Something could have gotten broken in the meantime. You don't have to do anything you don't want, I don't care if you don't test anything today, but you also cannot state as fact something that isn't really a fact that applies nowadays, many WHDLoad versions later, because it only adds noise to the issue. Going with what Zetro said above there is no way you have used the card with the *exact same configuration* because we don't have the exact same configuration (WHDLoad + MX1230 is too vague)

[edit] Zetro, you're a good fella but I think you are coming down too hard on me for no good reason, probably by misinterpreting what I wrote. That is all what I want to say here. I have no problem with you and hopefully you understand better what I meant to say before.

[Zetr0]

@Akira

My friend its not about you, its about the process - write your efforts up here, don't expect someone to "do it for you" there is lots you can do before you offer it to Wepl - theres a few guys in here that will be willing to help.

I am not being "aggressive" with you, perhaps you mistake my exacerbation at your current method as such - it isn't I am just trying to relay the tools you need to do this yourself and to help others to help you.

While I don't have the adaptor card in question, I have had plenty of experience and a nice collection of hardware that I am more than happy to offer help and apply to this problem - but to do this I need data to work with

I find this problem "intriguing" and I suspect that perhaps it may be something errant in the 030 on board the card - for that I need mask revisions and identical test set ups to see if it is either a peculiarity with the card or possible installed CPU's / FPU's - again this cannot be done without multiple instances of data.

So rather than getting aggravated (which is not how to solve problems) - lets get on the same page - put together a list of say 10 games that fail. put together a simple HDF with OS3.1 and the latest WHDLoad and test on the host - then others with the card can do the same. Then we measure the results - modify a tiny bit then test again and again and again....

small steps -

if you want help putting together a HDF for everyone with one of these cards to test at home let me know.

[jotd]

I had a mbx1230 20 years ago and I already had problems with games at the time. Bad quality from the starr I guess.

[Amiga1992]

Originally Posted by Zetr0

My friend its not about you, its about the process - write your efforts up here, don't expect someone to "do it for you" there is lots you can do before you offer it to Wepl - theres a few guys in here that will be willing to help.

But my gripe is that I have been documenting my testing as best as I could (not everyone is as technically savvy as you) for years on this thread and the other one I linked. The tech specs about teh CPU and stuff I posted on this thread two times already.
Ultimately I don't think this testing of mine is taking me anywhere or helping anyone. And precisely because you don't have the card it is that I feel you shouldn't bother with this matter, only the people who do can provide solid data and unfortunately this is as good as we seem to be getting, with me going into wild guesses and some other guy saying his tests of yonder are still valid today, many WHDLoad versions late.
Also I feel Wepl's input is paramount on the matter, as the WHDLoad dev. But without a card, I don't think he would be able to make decisive tests.

Sorry about the confusion anyway.

I'll try, when time permits, to make an HDF of this, if I can at all. The other data, I provided before, and I'll post it again before it gets buried deeper in threads:

Motherboard: 1D.1
CPU: SC414075RC50 (06C74N DEWQ9019)
FPU: XC68882RC50A (1C12R DENR9003)
RAM: 64MB 70ns
Kickstart: 3.0
Workbench: 3.0
File Systems tested: SFS, PFS3AIO, FAT95

WHDLoad version: 18.2.5735

The data from other people remains to be put forward in the same fashion, just my data alone cannot possibly help.

To be clear I had myriads of issues with WHDLoad from day one too. So not only this version has a problem, but I can trace back to version 14 at least when I first registered WHD.

[Amiga1992]

Well here is me again with this fucking accelerator and problems
I'm ready to throw it out to the garbage, but before that, I will try one last resort: recapping it.

There are 8 electrolytic capacitors on it and maybe, MAYBE? they are creating issues? I get a lot of times fast blinking LED and yellow screen on bootup now, which I "fix" by taking the thing out, reseating the processors and RAM and pop it back in. It makes as much sense as blowing into a fucking game console cartridge, but it seems to work most times.

Anyway, Lionheart was giving me nasty graphics corruption before when the accelerator was in. Again, IT IS NOT the RAM, I tested without RAM and same issues. I tested with another good working new stick f RAM that I keep as a backup, same issues. But like "magic", as soon as I remove the accelerator, Lionheart works JUST FINE.

Is there anything else I can try?? I read on a thread about the crystal connection being unstable (it does seem to pop off easily). My motherboard does not need fixes as far as I know.

[BarryB]

Mate, I replaced my M1230XA with a Blizzy 1230 MKIV @ 50MHz with same FPU, best thing I ever did, no more pissing about trying to get the M1230XA to play nicely with RAM or WHDLoad!!

[Amiga1992]

Originally Posted by BarryB

Mate, I replaced my M1230XA with a Blizzy 1230 MKIV @ 50MHz with same FPU, best thing I ever did, no more pissing about trying to get the M1230XA to play nicely with RAM or WHDLoad!!

I know, I am close to this happening, I think I would be fine with another accelerator, maybe an ACA.
But I am wiling to give it a last try, even if just to make sure everyone knows this is either a problem with the hardware or just a very shitty accelerator for WHDLoad.

Gurrag: do you recall what capacitors are needed for this job?

[Amiga1992]

Talking to myself here, the last remaining moron trying to use this stupid card in the planet? Maybe P

Anyway, Gurrag never showed up so I had a look at the capacitors and I read these values: 100µF, 6.3V, but now there's a doubt: the capacitors seems to be saying 885°C? Which I can't seem to find around.

I found these caps, but they are 105°C. What is this °C rating for anyway? Never paid attention to it.

Does anyone know if these caps I found would be a good replacement for this board?

[zipper]

885°C - no way! Perhaps 85°C ? Max temperature rating. 105°C might have a longer life especially when cooling is suboptimal.

[Amiga1992]

Yeah i thought the same! Could it be B85°C? Let me take a photo of it with macro lens.

Here:



It says B85°C then.
Would these 105°C be OK then? I guess they would actually be better.
Does "low ESR" matter?

[Jope]

As a rule, you should replace like for like. If the original cap's datasheet doesn't say low ESR, you don't need low ESR. :-)

[Amiga1992]

Originally Posted by Jope

As a rule, you should replace like for like. If the original cap's datasheet doesn't say low ESR, you don't need low ESR. :-)

Yeah but where would I get "the original cap's datasheet"? I have no idea what this cap is. Is "IC" or "lc" a brand? I seriously have no idea how to find that.
Also, supposedly, low ESR (which basically is all I can find now) is "better", so would there be any harm in "upgrading"?