@FuZion

Keep saving..
Then download the Envy24HT ahi drivers from OS4Depot when you get your card..
Install everything and enjoy!

@wegster

Quote:

Some interesting comments from the first link:

I think the SBLive issue that happened on VIA-based PCs is a bit of a red herring here. The corruption and lockup problems only ever occured with machines that had both the 686 and a VIA north bridge. Systems with the 686 and a non-VIA NB (such as the AMD 750) were okay.

That's not to say it's an impossibility. The Articia S may be reacting as badly to the 686B's problems as the VIA north bridges.

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

@DrBombcrater

Quote:

DrBombcrater wrote: @wegster Quote: Some interesting comments from the first link: I think the SBLive issue that happened on VIA-based PCs is a bit of a red herring here. The corruption and lockup problems only ever occured with machines that had both the 686 and a VIA north bridge. Systems with the 686 and a non-VIA NB (such as the AMD 750) were okay. That's not to say it's an impossibility. The Articia S may be reacting as badly to the 686B's problems as the VIA north bridges.

Heh, were your ears ringing earlier? (I was wondering if you were going to jump into this one )

I've read _most_ of the links I posted, but still remain unclear on a single problem, or solution. AS you said, it appeared the problem was perhaps several, when combined.

It would seem (very) odd, though, that there have now been 2 users who were able to reproduce the crash, yet no longer can after swapping out their SB cards, no?

If you have any more insight or thoughts on this one, I would _love_ to hear them? (not just me, I'm sure most of us would! )

_________________
Are we not done with the same silly arguments and flames yet??!

@wegster

Quote:

(I seem to recall over and over about hearing about 'can't get specs from Via,' what was needed specifically? Maybe we'll get lucky and be able to dig this out from someone involved in the 686B/SB issues in the past.

AFAIR, this was in relation to:

a) the AC97 sound present on the rev 1 XE boards (nobody could work out how to initialise it)

b) UDMA for the on-board VIA IDE. With DMA active, the VIA sent loads of packets across the PCI bus.

Sounds like the same thing as this:
Quote:

Liveware seems to permanently transfer null-bytes over the PCI bus to the SB Live, even at times when it does not have to play notes.

Quote:

Deactivating SB Live's MIDI functions stops this behavior.

That sounds like one of the easier fixes, that DaveAE can probably put in the driver.

Quote:

When Creative designed the SB Live, they didn't design it to support PCI Bus mastering

Is there any way to switch off bus mastering for individual PCI slots?

Quote:

PCI Delay Transaction: 0 or off PCI Master Read Caching: 0 or off PCI Latency: any value from 0 to 32

That sounds like something that can be fixed within U-Boot, probably an Articia configuration thing, which ties up with your other information that says it is a northbridge configuration conflict.

Apologies for quoting so much, but this all looks like good info that deserves to be read again

Chris

_________________
"Miracles we do at once, the impossible takes a little longer" - AJS on Hyperion
Avatar is Tabitha by Eric W Schwartz

@Steff

Quote:

Could this be why I'm not making any headway with the midi?

Yes. Please send me an email on info@audio-evolution.com so I can send you a new driver.

_________________
Audio Evolution
http://www.audio-evolution.com

@wegster

Quote:

Anyone have an updated list of non SB/Creative cards supported? I may just try that to verify the results.

intuitionbase has all supported cards.

_________________
Audio Evolution
http://www.audio-evolution.com

@FuZion

Quote:

Can I just clarify what I am reading, make sure I'm not getting the wrong end of the stick? The package: "driver/audio/envy24ht_ahi.lha" on OS4Depot, happens to support the SBLive & it appears that this driver, so far, has begun to prove more stable?

No, the envy24ht driver supports cards with Envy24HT chips and the emu10kx driver supports cards with emu10kx chips, just like you can't use ATI drivers on nVidia cards.

_________________
Audio Evolution
http://www.audio-evolution.com

@wegster


Quote:

Although DaveAE says there's some work to be done yet on the onboard chipset driver, so it's _possible_ that's a different issue..but still would need a resolution.

No no, not regarding playback, only volume control and input selection. Playing back is practically the same and very simple for all drivers:
1) IRQ
2) Mix
3) Do endian conversion on mix buffer and copy to DMA buffer
4) Flush caches

_________________
Audio Evolution
http://www.audio-evolution.com

@wegster

Quote:

DaveAE- any of this doing anything for ya?

Well, I've known for a long time that the SB Live! is a bitch, certainly together with the VIA686, but I can't tell if it's hogging the PCI bus on the A1. Anyway, since the microA1 has problems too with its CMI chip, I think there is more to it than this.

_________________
Audio Evolution
http://www.audio-evolution.com

@wegster

Quote:

Umm, so does anyone have this info to be corrected in uBoot? It would appear from empirical testing the Via referenced driver patch did NOT cure the problem, but it's possible the initialization issue might.

No, we and MAI tried to get this info from VIA, but we didn't get the info since we are nobodies to them.

_________________
Audio Evolution
http://www.audio-evolution.com

@Ivan

Quote:

I'll try tinkering with the newer drivers on os4depot since i read it works with the sblive (god knows why eh?). But later, im right tuckered out.

There are no newer emu10kx drivers, sorry.

_________________
Audio Evolution
http://www.audio-evolution.com

@Samwel

Quote:

Why can't VIA give Hyperion the docs they need on 686B.. Aaaargh!

We have docs on the 686B, we just don't know what they and/or Creative f*cked up.

_________________
Audio Evolution
http://www.audio-evolution.com

@DaveAE

Quote:

DaveAE wrote: @wegster Quote: DaveAE- any of this doing anything for ya? Well, I've known for a long time that the SB Live! is a bitch, certainly together with the VIA686, but I can't tell if it's hogging the PCI bus on the A1. Anyway, since the microA1 has problems too with its CMI chip, I think there is more to it than this.

But you think they're the same core problem, or different? The uA1 CMI chipset doesn't use the emu10k driver, correct (or core code?)

Thanks for the responses. My Terratec will ship out to me Monday, expecting Wed or Thursday hopefully, so we'll at least have another data point to see if my problems go away as well.

I re-read some of the 686b/SB pages, mostly from The 686B page and am wondering about the section where he's discussing resetting specific register values (there's more info, no one do this if they're even able to, but a summary of the changes follow below). Can any of this be tried, perhaps via uBoot if it ultimately corrects the problem, or can someone comment on the seeming validity of this?

What about disabling MIDI as a test driver to see if that has any effect as it was mentioned as a possible 'fix' as well?

Quote:

6) Now that you have seen the screen let me explain briefly what the registers we are looking at are, these are not the order in which I make the changes, just an explanation of how these registers alter your system: 52 (HEX FB on the example above) – The main reason this register is of interest, to us is to allow us to remove “Disconnect enable when STPGNT detected”. STPGNT means “Stop Grant”, and what this function does is allow the AMD processor to stop for a “pause” when it isn’t doing anything. This really has no other function (as far as I can see) than helping keep the AMD processor cool. However – as a last step in the fixing process this can sometimes be the best way to fix a bug. Only problem is, you need to make sure you have good cooling for your Processor, and case cooling is also a good idea – but lets look at the others before we change anything (again I use this as a last resort) – usually my HEX here is 7B but I changed it for this example. 64,65,66,67 (HEX E4, E4, D4, C4 on the example above) are all the locations for memory timing. Some BIOS’ appear to be getting these figures really wrong when left to its own resource (don’t ask why – I don’t have a clue). I have two memory chips – both are running at 2-way-interleave (the most stable for this system). Now the way to make sure which are the “active” memory locations is by looking at 58, and 59 (circled red in example 2). Where there is a number means there are memory chips in that memory socket – a 64 or a 128 Meg chip is the Hex number 8. Therefore as you can see from my system that I have two 64 or 128 Meg chips in slot 1 and slot 2, and there is nothing in slot 3 and 4. So therefore the only locations I will alter are offset 64 and 65, which read E4 and E4. If there was memory in Slot 3 it would be 66, or slot 4 would be 67. 70 (HEX CE on my example) – this is something I try not to change, but honestly sometimes it is the only way to get things working. Firstly if you use this fix you are bound to have to use the offset 52 fix also to stop major crackling on your SBLive (or other soundcard). Not everyone has had this problem, but a lot have. What we are doing here is turning off some advanced features of the Northbridge; however in the case of the SBLive some of these features really help complicate issues that card has with Bus Mastering. 71 (HEX 08 on my example) is something I change on occasion, main reason for this is because in turning this off you are disabling some other advanced functions of the Northbridge. 75 (HEX 81 on my example) allows us to manage the latency, or the “bus master time out” to stop the some corruption, giving the CPU and Northbridge some time to handing the information better. In fact one of the first fixes I got working for the 686b issue in Win98SE was to just change this offset to 83 and it fixed almost everything that was being caused by the so called “686b bug”. However this didn’t work alone in Win2k, but it was my first attempt at trying out just one change with George’s guidance. 76 (HEX 52 on my example) is what VIA Ltd came out with to fix the problems many people were having with their computers and corruption of data. The fix from VIA however doesn’t work on all systems, this is because the patch doesn’t change some systems with soundcards which are not Creative Labs SBLive (generation 1 and 2). Therefore anyone with a Creative Labs SBLive 5.1 will not get the patch working for him or her. Also there have been some people who haven’t had the patch work for them even if they have a first generation SBLive card – so far I can’t seem to see why this isn’t happening. This is the first fix I make sure is set up on all systems with the KT133 and KT133a/e boards, mainly because the seems to fix all issues with Southbridge corrupted data.

_________________
Are we not done with the same silly arguments and flames yet??!

@wegster

All those 'fixes' listed concern register locations on the KT133(A) North Bridge, so they have no relevance at all to the A1. The specific functions controlled by those registers may be present on the Articia S (but by no means all, the Articia isn't as tweakable as most x86 type designs) but the hardware behind them is different so even figuring out how to do the same things on the Articia probably wouldn't help.

I still believe that the SBLive is not really a factor and that either the 686B's effect on the PCI bus has far more widespread effects than just the UDMA/ethernet lockup, or the Articia's PCI controller is not performing quite as it should.

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

@moderators

would you please rename this thread to
something like

'coding audio for OS4'

the new drivers and supported cards
indicate a healthy audio system in the
works, but the title of this ongoing
thread implies the opposite...

thanx

@glokraw
Quote:

would you please rename this thread to something like 'coding audio for OS4' the new drivers and supported cards indicate a healthy audio system in the works, but the title of this ongoing thread implies the opposite...

Or it could be left the way it is, as most people know the thread as "Audio is STILL crashing OS4".

Also, this title still applies to many a system out there, including mine...

_________________
After a decade away from the scene, I am back!

@DWolfman

BTW, I did get to try this out on my XE system. I could not get it to hang even once. I tried with both AmigaAMP (using the WarpUP decoder) and Amplifier (using the OS4 mpega.library), while transferring the YDL 3 ISO images across my network to my server. No problems whatsoever.

I have my drives on a Sii0680 card, and the sound card is a Soundblaster 16 (works with the SB128 driver), so I know it is not one of the "Live" cards or even using the EMU10K chipset.

Last edited by DWolfman on 10-Jul-2005 at 01:30 AM.

_________________
This posting, in it's entirety, is the opinion and/or statement of the author and does not reflect the views and/or position of this site.

@DrBombcrater
Thanks for that info, I didn't realize that.

Not sure if that's the case for the rest of the fixes, though. I guess I'm hoping for an OS4 dev to jump in after looking at the available info and see what might pertain to our A1s.

_________________
Are we not done with the same silly arguments and flames yet??!

@DWolfman
Wanna buy an SB Live!, cheap?

So, is there ANYONE _with_ AN SB Live/emu10k chipset that does NOT have this problem?

(confirmations of working configs still needed as well..)

_________________
Are we not done with the same silly arguments and flames yet??!

@wegster

Quote:

Wanna buy an SB Live!, cheap?

Heh, no thanks! Already have one of those in the kids' PC. Only reason it's working fine there is that board doesn't have a VIA 686b southbridge.

Quote:

So, is there ANYONE _with_ AN SB Live/emu10k chipset that does NOT have this problem? (confirmations of working configs still needed as well..)

And THAT is the $6000000 question.

_________________
This posting, in it's entirety, is the opinion and/or statement of the author and does not reflect the views and/or position of this site.