@mlehto

Quote:

This is public channel, open for everyone to check, how is Amiga today. Not good, if he/she see only troubles around.

Pssst let me tell you something. There are some troubles. I know too much complaining and all scares of people. But hey like you said it's a public channel and that means also a place to talk about problems. Or in this case lack of communication.

Coder

_________________
Can't get enough of me?
The Bucket Diary Blog
The Bucket Diary Twitter Account

@Rudei

Quote:

What? No fixers ever? Why not?

I agree with DrBombcrater. There was never any word about it. All we are waiting for is the official word. YES/NO. And my money is on NO.

Coder

_________________
Can't get enough of me?
The Bucket Diary Blog
The Bucket Diary Twitter Account

@mlehto

Quote:

This is for others also, if you find working hardware, please share information with others :)

We do!

At least try to, at AHS

_________________
Site admins are people too..pooff!

@Coder

>Pssst let me tell you something. There are some troubles. I know too much complaining and all scares of people. But hey like you said it's a public channel and that means also a place to talk about problems. Or in this case lack of communication.

I definetly know lot about problems, since I have SE myself.

Writed here, pointed everywhere and specially for people, who brings problems up without owning device, bringing up rumours about problems w/o any facts etc. Sorry, it wasn't nothing to do with you. Just didn't want to point anyone specially.

Discussion about known problems is different, but I made it too abstract then, if you (and maybe others) misunderstood it. Sorry about it.


I'm not moderator for sure, but I think, that some kind of writing hurts whole A1... thats it.

Miikka

Last edited by mlehto on 08-Jan-2005 at 07:42 AM.

@mlehto

Quote:

Writed here, pointed everywhere and specially for people, who brings problems up without owning device, bringing up rumours about problems w/o any facts etc.

Point taken. You are rigth, that is something we don't need.

Quote:

Sorry about it.

Don't be. Was not directed at you and I got you wrong about the people that even don't own one. It was more about the fact that we can't even say that we want an answer.

Coder

_________________
Can't get enough of me?
The Bucket Diary Blog
The Bucket Diary Twitter Account

@Coder

Quote:

>>Sorry about it.

>Don't be. Was not directed at you and I got you wrong about the people that even don't own one. It was more about the fact that we can't even say that we want an answer.

All fine :)

Here is allready introduced some solutions. Alltought they don't repair our mobo's. Cheap UCHI card is one (posible, not sure or secure) solution, if you allready have sil0680. If you don't, there is allso combo-cards. They are expensive, but if you think that you need one, then it is ok.

These are anyway interim solutions, I still wait some clarification how to fix SE mobo's. There is possibility, that fixes are similar ... Ole-Egil ??

Anyway, somebody started this discussion and it gave for us some fruits, so it was more than fine.

Miikka

@Coder

Ok, so you want to know if there will be a fix for the SE. How about this answer:
Don't know yet.

Seriously:
Very likely not. Your SE is probably suffering from a lot more than just what those lucky buggers with XEs are complaining about. I've written mine off as a loss already, I'm only keeping it for sentimental value. But I was lucky enough to buy it at a VERY good price, so I most certainly won't be complaining.

Let's just say that I understand your frustration, but AT THIS MOMENT, more whining won't make a licking difference (towards a positive solution, anyway).

I keep forgetting, are you using an SIL for UDMA?

There seems to be something VERY wrong with the onboard USB on the SE, that makes storage devices fail utterly. And I think there is WAY too little interest in doing a lot of research into finding the problem with these boards. If you're lucky, you might get a discount on a PCI USB card from Eyetech at some point. But at this point it's too early to tell.

I also think it would be a BAD call from Eyetechs side to withdraw all SE/XE boards, fix them and reship them now, because who knows how many OTHER small itty-bitty bugs there are that means extra rounds...

Until OS4 is COMPLETELY finished and everyone agrees that the most up-to-date set of patches is definetly what everyone wants to have on their AmigaOnes, I would advise everyone against jumping on the bandwagon here. It might just be for nothing, ending up with another trip home for all those boards

I just KNOW you would groan a bit if you got your board back from your dealer after being without it for 2 months, only to find that last weekend someone discovered another major bug that needs patching, leaving you without a board for another 2 months

There's a time for everything, and _one day after 'When it's done(TM)"_ seems like the best time for most of it. In my quite humble opinion, of course. I respect it if you disagree

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@mlehto

I believe the USB debounce thingie is identical, but until someone can tell me excactly which capacitors should be replaced with resistors there's not a ####ing lot I can do other than guess.

That's MY gripe with the whole situation. I am COMPLETELY unwilling to take the "Unless you want to solder itty bitty components yourself this isn't for you"-responses as a no. Yes, I have hand-soldered 0402 SMDs previously. I designed my own PCB last year that uses several of them. And a HUGE bunch of 0603 SMDs. What is the problem? Just TELL ME WHAT TO LOOK FOR!

The XE has room for a programmable logic circuit which the SE doesn't have, so it COULD be that the "Ethernet+VIA DMA isn't possible" fix has something to do with this, but I'm just guessing here. Again, noone wants to tell me the whole story, leaving me to speculate. Badly

Hmm, I'm hungry. I must be the experiencing the dreaded ArticiaS DMA bug. Or just low blood sugar. Whatever

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

Quote:

I just KNOW you would groan a bit if you got your board back from your dealer after being without it for 2 months, only to find that last weekend someone discovered another major bug that needs patching, leaving you without a board for another 2 months

I'm going to be real pissed if my board is away for 2 months. That's to frigging long.

That's why it's important that the fixes are shared with dealers in every country so they can find a person or to assertain if they can do the fixes themself. Thats why I would also like to assertain myself if I can do the fixes. I'm not an electronics nitwitt so I can decide if I can or can not do the fixes.

If I can't then I would like to have it fixed in the Netherlands. In that case I will make an apointment and drive to it, get it fixed, tested and return home. Thats how this fix should happen.

_________________
AmigaOne-XE G3 OS 4.
A4000 PPC
A1200 PPC

@olegil

>I believe the USB debounce thingie is identical, but until someone can tell me excactly which capacitors should be replaced with resistors there's not a ####ing lot I can do other than guess.


Yes, I am after them also, but without luck yet. I somehow presume, that USB-thingy can be identical with XE, but also DMA can be more complicated. Let's see, what happens in future.




>That's MY gripe with the whole situation. I am COMPLETELY unwilling to take the "Unless you want to solder itty bitty components yourself this isn't for you"-responses as a no. Yes, I have hand-soldered 0402 SMDs previously. I designed my own PCB last year that uses several of them. And a HUGE bunch of 0603 SMDs. What is the problem? Just TELL ME WHAT TO LOOK FOR!


No worry, I know you as others, so it might not be a broblem... :) I don't want to make it myself, but my uncle have some kind of electronics factory, so I guess, that they have lab also.




>The XE has room for a programmable logic circuit which the SE doesn't have, so it COULD be that the "Ethernet+VIA DMA isn't possible" fix has something to do with this, but I'm just guessing here. Again, noone wants to tell me the whole story, leaving me to speculate. Badly


Seems to be like that just now. I thought, that you may know it ... :(

Too little info avail just now.

Is that programmable chip something to do with cache-coherency situation ?? If so, there would be some hope to get SE boards fixed...

Maybe our swedish friends knows it better ... ? There is also this NDA situation, if this info is "classified" and not public yet...

Seems, that future of our SE-boards are hands of Alan by now.




>Hmm, I'm hungry. I must be the experiencing the dreaded ArticiaS DMA bug. Or just low blood sugar. Whatever


Remember eat something every now and then ... :)

Miikka

@olegil

Quote:

Very likely not. Your SE is probably suffering from a lot more than just what those lucky buggers with XEs are complaining about. I've written mine off as a loss already, I'm only keeping it for sentimental value. But I was lucky enough to buy it at a VERY good price, so I most certainly won't be complaining.

I know. I did not had it at a VERY good price. Normal as everyone else.

Quote:

I keep forgetting, are you using an SIL for UDMA?

Nope. I am waiting with that. You were talking about a PCI USB card. Well that would be cool if that works. Sure with those 2 cards and soundcard it becomes full. And yes waiting for the final release is a good thing. That's why I did not buy the SIL.

Quote:

I just KNOW you would groan a bit if you got your board back from your dealer after being without it for 2 months, only to find that last weekend someone discovered another major bug that needs patching, leaving you without a board for another 2 months

I would.

At least E-UAE works great. The sound works good with that and not with anything else. I know the boards were beta but you see in the end it comes down to this for me, I gave my support back then so I expect Eyetech to give some back. Not in money or whatever but in replies. That's all I ask for. Having someone talking to me that tells me what is going on. Ok maybe not now yet but certainly with the final release out. That's all I ask, really.

Hmm maybe I am complaining because I still don't have a new coffee machine. My Cup of Joe electric blew the bottom out of the glass can.

Coder

_________________
Can't get enough of me?
The Bucket Diary Blog
The Bucket Diary Twitter Account

@olegil

Quote:

Very likely not. Your SE is probably suffering from a lot more than just what those lucky buggers with XEs are complaining about.

Bet on it. There's some serious issue with the PCI bus on the SE that is causing problems for every PCI device in the system, card or board-mounted. I doubt MAI even have the skills to find out what it is, so I'm not holding my breath for a fix.

Quote:

If you're lucky, you might get a discount on a PCI USB card from Eyetech at some point.

That would be nice, if fitting a PCI USB card actually cured the USB issues. But it doesn't. Debounce is fixed, but mass-storage still dies with its legs in the air, exactly the way it does with the on-board controller.

Quote:

Until OS4 is COMPLETELY finished and everyone agrees that the most up-to-date set of patches is definetly what everyone wants to have on their AmigaOnes, I would advise everyone against jumping on the bandwagon here. It might just be for nothing, ending up with another trip home for all those boards

The snag with that idea is that when OS4 is done, possibly many months from now, Eyetech will be even less likely than they are now to pay the slightest bit of attention to the SE. They seem to regard people who bought SEs as mugs who don't deserve to have their concerns acknowledged, far less addressed.

Being treated like that really makes be very angry indeed.

And it is just plain bad business to treat your customers in such a dismissive way. There must be over a hundred people with SEs, not an insignificant number in this market. When they go shopping for another OS4 machine, how many will be willing to place themselves in Eyetech's hands again?

I hope that new PPC750-based A1200 busboard is a solid product, because I can see one of those in my future...

_________________
Who do you serve, and who do you trust? - Galen

@olegil


Sorry to hear, that SE of yours is next to unusable ...

Overall, thank you about clarifications, they made some sense for this SE mobo situations.

>I also think it would be a BAD call from Eyetechs side to withdraw all SE/XE boards, fix them and reship them now, because who knows how many OTHER small itty-bitty bugs there are that means extra rounds...


I will gladly change my SE to XC, when they come avail :) If I have money for that in time.



Miikka

Last edited by mlehto on 08-Jan-2005 at 12:36 PM.

@DrBombcrater

>>Very likely not. Your SE is probably suffering from a lot more than just what those lucky buggers with XEs are complaining about.


>Bet on it. There's some serious issue with the PCI bus on the SE that is causing problems for every PCI device in the system, card or board-mounted. I doubt MAI even have the skills to find out what it is, so I'm not holding my breath for a fix.


How this PCI-bus problem shows itself up ??




>>If you're lucky, you might get a discount on a PCI USB card from Eyetech at some point.

>That would be nice, if fitting a PCI USB card actually cured the USB issues. But it doesn't. Debounce is fixed, but mass-storage still dies with its legs in the air, exactly the way it does with the on-board controller.


I'm sorry about that ... :( are you sure, that it is not just because usb is under development ??


Miikka

@DrBombcrater

Quote:

That would be nice, if fitting a PCI USB card actually cured the USB issues. But it doesn't. Debounce is fixed, but mass-storage still dies with its legs in the air, exactly the way it does with the on-board controller.

####!

Quote:

I hope that new PPC750-based A1200 busboard is a solid product, because I can see one of those in my future...

That certainly is an option. I do hope we will see other vendors bringing AmigaOne type computers on the market.

Coder

_________________
Can't get enough of me?
The Bucket Diary Blog
The Bucket Diary Twitter Account

@mlehto

Quote:

How this PCI-bus problem shows itself up ??

When a driver/device pair that's known to work on the XE fails or doesn't work quite as intended on the SE.

Soundblaster 128 and PCI USB cards don't work correctly on the SE but do, apparently, run fine on the XE using the same drivers. And in some cases the SE's ethernet shows up problems (like dying during large file transfers over LANs) that don't happen on the XE, again using the same driver and chip.

I'm reaching a bit here, in calling these symptoms of a PCI bus problem. There may just be wierd issues in the SB128 and USB drivers that cause them to fail that way only on the SE, and the ethernet lock-ups may be down to layout problems on the SE's PCB.

But I really do think something is broken -- that's why I bought a PCI USB card to test with when SE's mass-storage problem surfaced. I was rather convinced that if the SE's undiagnosed problem messed up a VIA USB controller on the motherboard it would do the same thing to a VIA USB controller mounted in a PCI slot.

There's still some hope it's just a glitch in the USB driver, but experience shows that with the A1 the worst-case scenario is usually the right one.

_________________
Who do you serve, and who do you trust? - Galen

@DrBombcrater

Quote:

>>How this PCI-bus problem shows itself up ??


>When a driver/device pair that's known to work on the XE fails or doesn't work quite as intended on the SE.

>Soundblaster 128 and PCI USB cards don't work correctly on the SE but do, apparently, run fine on the XE using the same drivers. And in some cases the SE's ethernet shows up problems (like dying during large file transfers over LANs) that don't happen on the XE, again using the same driver and chip.

Yes, it seems to be and it makes me litle bit nerve about PCI of SE. I am personally never met this eth problem (after sil). Only dying, not crc errors ?? That one I can test myself.

Very stupid question. Are you tried to drop FSB to 100MHz. I mean, that if something reach it's limits in mobo, then speed dropping may help. In case there is not enought marginals for speed/timings in some bus/busses, or speed causes some signal bouncing (sorry, my head, I forgot correct name for this...:) ... PCB-layout engineering, components ... I don't continue, it is something, what anyone of us can't help.

And very brobably yor SE was allready set to 600 MHz with proper cooling ? If it runs fine at 666 MHz, it may be better to drop down during testing. I know, that these are stupid, but ...

I changed my cooler to zallman (it runs stable in 666MHz now...), and it works well since then. Somehow I have overall feeling, that this mobo is very sensitive. My dealer said, that his clients experience about SE-mobos vary quite much. Someone have all probs, other don't have them at all(I have them all). It is STRANGE... These boards are from 3rd production round, mine is from 1st public. Boards may vary from round to round (?).

It may be allso question of power-supply. I calculate in my head, how much this SE will take power. I have Ati7000/64 Mb, 512 Mb kingston, 1 maxtor80Gb, sil0680, CDROM, CDRW and thats it. I think that it is maybe 70-100W. Not much more than 100W anyway, I think. Then my PSU is 350W. So I doubt, that it works properly with this load. So it may better to test with some 150W PSU and see, what happens. But I'm not the richest man around here, so I have to wait.


>I'm reaching a bit here, in calling these symptoms of a PCI bus problem. There may just be wierd issues in the SB128 and USB drivers that cause them to fail that way only on the SE, and the ethernet lock-ups may be down to layout problems on the SE's PCB.

Yes, all devices (in question) are anyway in PCI-bus.


>But I really do think something is broken -- that's why I bought a PCI USB card to test with when SE's mass-storage problem surfaced. I was rather convinced that if the SE's undiagnosed problem messed up a VIA USB controller on the motherboard it would do the same thing to a VIA USB controller mounted in a PCI slot.

Too bad, other cross problem. Same like adding ethernet card won't help.


>There's still some hope it's just a glitch in the USB driver, but experience shows that with the A1 the worst-case scenario is usually the right one.

It could be. It is very weird anyway.

Did you or do you oppoturnity to cross-check your PCI USB card and USB-devices with XE somewhere? I know, that someone maybe said, that these devices works, but it is allways better to check them yourself.

Hope that helps some ... Anyway, will you keep us others informed, if you get some new information or you found solution howto get USB to work ? Thank you :)

I don't have mobo's manuals in hand. If I remember correctly, possibilities to adjust bus-speeds in mobo are very limited. Anyway you could test 100MHz FSB, if it helps.


Miikka

Last edited by mlehto on 08-Jan-2005 at 01:50 PM.
Last edited by mlehto on 08-Jan-2005 at 01:47 PM.

@DrBombcrater

Quote:

Soundblaster 128 and PCI USB cards don't work correctly on the SE but do, apparently, run fine on the XE using the same drivers. And in some cases the SE's ethernet shows up problems (like dying during large file transfers over LANs) that don't happen on the XE, again using the same driver and chip.

I think it is more related to a driver problem. With my SE, my SB128 works fine under Debian but not under AOS4. And I haven't had problems with file transfers or connection dying when using ethernet with my LAN or internet (both under Debian and AOS4). And since the first installation, UDMA is active under Debian (if not, chances are that boot will fail or several error mesages appears in console). So I guess, I HOPE, it is a driver issue. Don't know about USB though, not tried it yet.

_________________
PowerBook 5.2 MorphOS 3.15
PowerBook 5.8 MorphOS 3.15
Amiga 1200 BPPC/BVision AOS4.1 FE

@tonyw

Quote:

The USB "issue" is this: Some XEs (and perhaps SEs, but I don't know about them) were fitted with capacitors instead of pullup resistors on the USB signal lines (two for each of the four USB ports). The effect is that when a device is unplugged, its absence is not noted and so it is not detached. It may be that this is why some mass-storage devices are not detected unless they are plugged in before the power is turned on, but I don't know for sure. Anyway, it does not affect things like mouses, keyboards, printers, all the stuff you are likely to use. The only time it can cause a problem is if you disconnect something, then reconnect it - it might not reappear, or might appear twice. On my XE, I can connect/disconnect any number of mouses and they all attach and detach properly, so I don't have a problem. If you do have a problem, the easy fix is to use an external USB hub (one of those $10 dual- or four-port hubs that you can get from any newsagent). That isolates the attach/detach action from the XE port and guarantees proper operation. That's all you need, and it even gives you additional flexibility - more ports!

This may true for XE, but certianly not SE. I haven't been able to use mass storage at all. My machine hangs 2-10 seconds after recognising, with or without hub. Even hub with external power supply didn't work!

This also happens in Linux just the same!!!

Firbodi

_________________
MicroA1 G3, 512MB RAM, DVD-RW; and A1-SE G3,512MB RAM, CD-RW & DVD-ROM
Avatar: Perspolis, Capital of Achaemenid Empire 500 B.C.

@emeck

Quote:

I think it is more related to a driver problem. With my SE, my SB128 works fine under Debian but not under AOS4.

The SB128 working under Linux just proves the problem is fixable. It doesn't give any clues to what's happening under OS4 -- why the same card and same driver, running under the same OS, with the same firmware and the same North Bridge, should work fine on the XE but bork horribly on the SE.

Perhaps the OS4 SB128 driver is trying something the Linux one doesn't, something the SE can't cope with. That would raise the uncomfortable question of how many future drivers will die on the SE because they do the same thing.

@mlehto
Quote:

I don't have mobo's manuals in hand. If I remember correctly, possibilities to adjust bus-speeds in mobo are very limited. Anyway you could test 100MHz FSB, if it helps.

Going to 100MHz doesn't help, nor does underclocking the CPU, changing the memory, moving PCI cards, or anything else I've tried.

_________________
Who do you serve, and who do you trust? - Galen